[AMDGPU] Enable amdgpu-sw-lower-lds pass in pipeline.

Author: skc7

llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp

@@ -343,6 +343,12 @@ static cl::opt<bool> EnableStructurizerWorkarounds(
     cl::desc("Enable workarounds for the StructurizeCFG pass"), cl::init(true),
     cl::Hidden);
 
+static cl::opt<bool>
+    EnableSwLowerLDS("amdgpu-enable-sw-lower-lds",
+                     cl::desc("Enable lowering of lds to global memory pass "
+                              "and asan instrument resulting IR."),
+                     cl::init(true), cl::Hidden);
+
 static cl::opt<bool, true> EnableLowerModuleLDS(
     "amdgpu-enable-lower-module-lds", cl::desc("Enable lower module lds pass"),
     cl::location(AMDGPUTargetMachine::EnableLowerModuleLDS), cl::init(true),
@@ -765,6 +771,8 @@ void AMDGPUTargetMachine::registerPassBuilderCallbacks(PassBuilder &PB) {
         // We want to support the -lto-partitions=N option as "best effort".
         // For that, we need to lower LDS earlier in the pipeline before the
         // module is partitioned for codegen.
+        if (EnableSwLowerLDS)
+          PM.addPass(AMDGPUSwLowerLDSPass(*this));
         if (EnableLowerModuleLDS)
           PM.addPass(AMDGPULowerModuleLDSPass(*this));
         if (EnableAMDGPUAttributor && Level != OptimizationLevel::O0)
@@ -1071,6 +1079,10 @@ void AMDGPUPassConfig::addIRPasses() {
   // Replace OpenCL enqueued block function pointers with global variables.
   addPass(createAMDGPUOpenCLEnqueuedBlockLoweringPass());
 
+  // Lower LDS accesses to global memory pass if address sanitizer is enabled.
+  if (EnableSwLowerLDS)
+    addPass(createAMDGPUSwLowerLDSLegacyPass(&TM));
+
   // Runs before PromoteAlloca so the latter can account for function uses
   if (EnableLowerModuleLDS) {
     addPass(createAMDGPULowerModuleLDSLegacyPass(&TM));