Add a RequireStructuredCFG Field to TargetMachine.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@196634 91177308-0d34-0410-b5e6-96231b3b80d8

docs/WritingAnLLVMBackend.rst

@@ -238,6 +238,12 @@ For some targets, you also need to support the following methods:
 * ``getTargetLowering()``
 * ``getJITInfo()``
 
+Some architectures, such as GPUs, do not support jumping to an arbitrary
+program location and implement branching using masked execution and loop using
+special instructions around the loop body. In order to avoid CFG modifications
+that introduce irreducible control flow not handled by such hardware, a target
+must call `setRequiresStructuredCFG(true)` when being initialized.
+
 In addition, the ``XXXTargetMachine`` constructor should specify a
 ``TargetDescription`` string that determines the data layout for the target
 machine, including characteristics such as pointer size, alignment, and

include/llvm/Target/TargetMachine.h

@@ -88,6 +88,7 @@ protected: // Can only create subclasses.
   unsigned MCUseLoc : 1;
   unsigned MCUseCFI : 1;
   unsigned MCUseDwarfDirectory : 1;
+  unsigned RequireStructuredCFG : 1;
 
 public:
   virtual ~TargetMachine();
@@ -156,6 +157,9 @@ public:
     return 0;
   }
 
+  bool requiresStructuredCFG() const { return RequireStructuredCFG; }
+  void setRequiresStructuredCFG(bool Value) { RequireStructuredCFG = Value; }
+
   /// hasMCRelaxAll - Check whether all machine code instructions should be
   /// relaxed.
   bool hasMCRelaxAll() const { return MCRelaxAll; }

lib/CodeGen/BranchFolding.cpp

@@ -83,7 +83,11 @@ INITIALIZE_PASS(BranchFolderPass, "branch-folder",
 
 bool BranchFolderPass::runOnMachineFunction(MachineFunction &MF) {
   TargetPassConfig *PassConfig = &getAnalysis<TargetPassConfig>();
-  BranchFolder Folder(PassConfig->getEnableTailMerge(), /*CommonHoist=*/true);
+  // TailMerge can create jump into if branches that make CFG irreducible for
+  // HW that requires structurized CFG.
+  bool EnableTailMerge = !MF.getTarget().requiresStructuredCFG() &&
+      PassConfig->getEnableTailMerge();
+  BranchFolder Folder(EnableTailMerge, /*CommonHoist=*/true);
   return Folder.OptimizeFunction(MF,
                                  MF.getTarget().getInstrInfo(),
                                  MF.getTarget().getRegisterInfo(),

lib/CodeGen/MachineBasicBlock.cpp

@@ -677,6 +677,11 @@ MachineBasicBlock::SplitCriticalEdge(MachineBasicBlock *Succ, Pass *P) {
   MachineFunction *MF = getParent();
   DebugLoc dl;  // FIXME: this is nowhere
 
+  // Performance might be harmed on HW that implements branching using exec mask
+  // where both sides of the branches are always executed.
+  if (MF->getTarget().requiresStructuredCFG())
+    return NULL;
+
   // We may need to update this's terminator, but we can't do that if
   // AnalyzeBranch fails. If this uses a jump table, we won't touch it.
   const TargetInstrInfo *TII = MF->getTarget().getInstrInfo();

lib/CodeGen/Passes.cpp

@@ -725,7 +725,10 @@ void TargetPassConfig::addMachineLateOptimization() {
     printAndVerify("After BranchFolding");
 
   // Tail duplication.
-  if (addPass(&TailDuplicateID))
+  // Note that duplicating tail just increases code size and degrades
+  // performance for targets that require Structured Control Flow.
+  // In addition it can also make CFG irreducible. Thus we disable it.
+  if (!TM->requiresStructuredCFG() && addPass(&TailDuplicateID))
     printAndVerify("After TailDuplicate");
 
   // Copy propagation.

lib/Target/R600/AMDGPUTargetMachine.cpp

@@ -72,6 +72,7 @@ AMDGPUTargetMachine::AMDGPUTargetMachine(const Target &T, StringRef TT,
     InstrInfo.reset(new SIInstrInfo(*this));
     TLInfo.reset(new SITargetLowering(*this));
   }
+  setRequiresStructuredCFG(true);
   initAsmInfo();
 }
 

lib/Target/TargetMachine.cpp

@@ -55,6 +55,7 @@ TargetMachine::TargetMachine(const Target &T,
     MCUseLoc(true),
     MCUseCFI(true),
     MCUseDwarfDirectory(false),
+    RequireStructuredCFG(false),
     Options(Options) {
 }
 

test/CodeGen/R600/jump-address.ll

@@ -1,6 +1,6 @@
 ;RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s
 
-; CHECK: JUMP @3
+; CHECK: JUMP @6
 ; CHECK: EXPORT
 ; CHECK-NOT: EXPORT