[LoopUnrollPeel] Add case where we should forget the peeled loop from SCEV.

[fhahn]

Summary

The test case requires the peeled loop to be forgotten after peeling,
even though it does not have a parent. When called via the unroller,
SE->forgetTopmostLoop is also called, so the test case would also pass
without any SCEV invalidation, but peelLoop is exposed as utility
function. Also, in the test case, simplifyLoop will make changes,
removing the loop from SCEV, but it is better to not rely on this
behavior.

[mkazantsev]

LGTM, but use forgetTopmostLoop instead.

llvm/lib/Transforms/Utils/LoopUnrollPeel.cpp

668

The correct way of doing this is L->forgetTopmostLoop(). Please us it to avoid potential sneaky bugs with containing loops in the future.

[fhahn]

Use forgetTopmostLoop and get rid of getting the parent loop, as it is not
required any longer.

[fhahn]

In D58192#1397922, @fhahn wrote:

Use forgetTopmostLoop and get rid of getting the parent loop, as it is not
required any longer.

I'll keep the code to get the parent loop for simplifyLoop when committing.

llvm/lib/Transforms/Utils/LoopUnrollPeel.cpp

if (ExitWeight) {
     MDBuilder MDB(LatchBR->getContext());
     MDNode *WeightNode =
         HeaderIdx ? MDB.createBranchWeights(ExitWeight, BackEdgeWeight)
                   : MDB.createBranchWeights(BackEdgeWeight, ExitWeight);
     LatchBR->setMetadata(LLVMContext::MD_prof, WeightNode);
   }
 
   // If the loop is nested, we changed the parent loop, update SE.
-  if (Loop *ParentLoop = L->getParentLoop()) {
-    SE->forgetLoop(ParentLoop);
+  if (Loop *ParentLoop = L->getParentLoop())
+    L = ParentLoop;
 
-    // FIXME: Incrementally update loop-simplify
-    simplifyLoop(ParentLoop, DT, LI, SE, AC, PreserveLCSSA);
-  } else {
-    // FIXME: Incrementally update loop-simplify
-    simplifyLoop(L, DT, LI, SE, AC, PreserveLCSSA);
-  }
+  SE->forgetLoop(L);
 
   NumPeeled++;
 
   return true;
 }

llvm/test/Transforms/LoopUnroll/peel-loop-scev-invalidate.ll

+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt < %s -S -loop-unroll -unroll-force-peel-count=1 -verify-scev -verify-dom-info | FileCheck %s
+
+
+define void @test1(i32 %k) {
+; CHECK-LABEL: @test1(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br label [[FOR_BODY_PEEL_BEGIN:%.*]]
+; CHECK:       for.body.peel.begin:
+; CHECK-NEXT:    br label [[FOR_BODY_PEEL:%.*]]
+; CHECK:       for.body.peel:
+; CHECK-NEXT:    [[INC_PEEL:%.*]] = add nsw i32 0, 1
+; CHECK-NEXT:    [[CMP_PEEL:%.*]] = icmp ult i32 0, [[K:%.*]]
+; CHECK-NEXT:    br i1 [[CMP_PEEL]], label [[FOR_BODY_PEEL_NEXT:%.*]], label [[FOR_END:%.*]]
+; CHECK:       for.body.peel.next:
+; CHECK-NEXT:    br label [[FOR_BODY_PEEL_NEXT1:%.*]]
+; CHECK:       for.body.peel.next1:
+; CHECK-NEXT:    br label [[ENTRY_PEEL_NEWPH:%.*]]
+; CHECK:       entry.peel.newph:
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
+; CHECK:       for.body:
+; CHECK-NEXT:    [[I_05:%.*]] = phi i32 [ [[INC_PEEL]], [[ENTRY_PEEL_NEWPH]] ], [ [[INC:%.*]], [[FOR_BODY]] ]
+; CHECK-NEXT:    [[INC]] = add nuw nsw i32 [[I_05]], 1
+; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i32 [[I_05]], [[K]]
+; CHECK-NEXT:    br i1 [[CMP]], label [[FOR_BODY]], label [[FOR_END_LOOPEXIT:%.*]], !llvm.loop !0
+; CHECK:       for.end.loopexit:
+; CHECK-NEXT:    br label [[FOR_END]]
+; CHECK:       for.end:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %for.body
+
+for.body:
+  %i.05 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
+  %inc = add nsw i32 %i.05, 1
+  %cmp = icmp ult i32 %i.05, %k
+  br i1 %cmp, label %for.body, label %for.end
+
+for.end:
+  ret void
+}