[LoopUnroll] Handle certain PHIs in full unrolling properly

Summary:
When reconstructing the CFG of the loop after unrolling,
LoopUnroll could in some cases remove the phi operands of
loop-carried values instead of preserving them, resulting
in undef phi values after loop unrolling.

When doing this reconstruction, avoid removing incoming
phi values for phis in the successor blocks if the successor
is the block we are jumping to anyway.

Patch-by: ebevhan

Reviewers: fhahn, efriedma

Reviewed By: fhahn

Subscribers: bjope, lebedev.ri, zzheng, dmgreen, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D66334

llvm-svn: 369886

2 files changed, 112 insertions, 3 deletions

diff --git a/llvm/lib/Transforms/Utils/LoopUnroll.cpp b/llvm/lib/Transforms/Utils/LoopUnroll.cpp
index e39ade52371..4a1edb3700c 100644
--- a/llvm/lib/Transforms/Utils/LoopUnroll.cpp
+++ b/llvm/lib/Transforms/Utils/LoopUnroll.cpp
@@ -711,7 +711,7 @@ LoopUnrollResult llvm::UnrollLoop(Loop *L, UnrollLoopOptions ULO, LoopInfo *LI,
 
   auto setDest = [LoopExit, ContinueOnTrue](BasicBlock *Src, BasicBlock *Dest,
                                             ArrayRef<BasicBlock *> NextBlocks,
-                                            BasicBlock *CurrentHeader,
+                                            BasicBlock *BlockInLoop,
                                             bool NeedConditional) {
     auto *Term = cast<BranchInst>(Src->getTerminator());
     if (NeedConditional) {
@@ -723,7 +723,9 @@ LoopUnrollResult llvm::UnrollLoop(Loop *L, UnrollLoopOptions ULO, LoopInfo *LI,
       if (Dest != LoopExit) {
         BasicBlock *BB = Src;
         for (BasicBlock *Succ : successors(BB)) {
-          if (Succ == CurrentHeader)
+          // Preserve the incoming value from BB if we are jumping to the block
+          // in the current loop.
+          if (Succ == BlockInLoop)
             continue;
           for (PHINode &Phi : Succ->phis())
             Phi.removeIncomingValue(BB, false);
@@ -794,7 +796,7 @@ LoopUnrollResult llvm::UnrollLoop(Loop *L, UnrollLoopOptions ULO, LoopInfo *LI,
         // unconditional branch for some iterations.
         NeedConditional = false;
 
-      setDest(Headers[i], Dest, Headers, Headers[i], NeedConditional);
+      setDest(Headers[i], Dest, Headers, HeaderSucc[i], NeedConditional);
     }
 
     // Set up latches to branch to the new header in the unrolled iterations or
diff --git a/llvm/test/Transforms/LoopUnroll/unroll-header-exiting-with-phis.ll b/llvm/test/Transforms/LoopUnroll/unroll-header-exiting-with-phis.ll
new file mode 100644
index 00000000000..e1fba5f6689
--- /dev/null
+++ b/llvm/test/Transforms/LoopUnroll/unroll-header-exiting-with-phis.ll
@@ -0,0 +1,107 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt < %s -S -loop-unroll -unroll-allow-partial | FileCheck %s
+
+; The phi which acts as input to func should not be undef. It should
+; have its loop-carried value (the load in for.cond) replaced accordingly
+; after unrolling the loop.
+
+define i16 @full_unroll(i16* %A) {
+; CHECK-LABEL: @full_unroll(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br label [[FOR_COND:%.*]]
+; CHECK:       for.cond:
+; CHECK-NEXT:    [[TMP2:%.*]] = load i16, i16* [[A:%.*]]
+; CHECK-NEXT:    br label [[FOR_COND_CLEANUP3:%.*]]
+; CHECK:       for.cond.cleanup:
+; CHECK-NEXT:    [[DOTLCSSA10_LCSSA:%.*]] = phi i16 [ [[TMP2_2:%.*]], [[FOR_COND_CLEANUP3_2:%.*]] ]
+; CHECK-NEXT:    [[TMP3:%.*]] = call i16 @func(i16 [[DOTLCSSA10_LCSSA]])
+; CHECK-NEXT:    ret i16 0
+; CHECK:       for.cond.cleanup3:
+; CHECK-NEXT:    [[PTR_1:%.*]] = getelementptr inbounds i16, i16* [[A]], i64 1
+; CHECK-NEXT:    [[TMP2_1:%.*]] = load i16, i16* [[PTR_1]]
+; CHECK-NEXT:    br label [[FOR_COND_CLEANUP3_1:%.*]]
+; CHECK:       for.cond.cleanup3.1:
+; CHECK-NEXT:    [[PTR_2:%.*]] = getelementptr inbounds i16, i16* [[A]], i64 2
+; CHECK-NEXT:    [[TMP2_2]] = load i16, i16* [[PTR_2]]
+; CHECK-NEXT:    br label [[FOR_COND_CLEANUP3_2]]
+; CHECK:       for.cond.cleanup3.2:
+; CHECK-NEXT:    [[PTR_3:%.*]] = getelementptr inbounds i16, i16* [[A]], i64 3
+; CHECK-NEXT:    [[TMP2_3:%.*]] = load i16, i16* [[PTR_3]]
+; CHECK-NEXT:    br i1 false, label [[FOR_COND_CLEANUP3_3:%.*]], label [[FOR_COND_CLEANUP:%.*]]
+; CHECK:       for.cond.cleanup3.3:
+; CHECK-NEXT:    unreachable
+;
+entry:
+  br label %for.cond
+
+for.cond:                                         ; preds = %for.cond.cleanup3, %entry
+  %.lcssa10 = phi i16 [ 123, %entry ], [ %.lcssa, %for.cond.cleanup3 ]
+  %i.0 = phi i64 [ 0, %entry ], [ %inc9, %for.cond.cleanup3 ]
+  %ptr = getelementptr inbounds i16, i16* %A, i64 %i.0
+  %tmp2 = load i16, i16* %ptr
+  %cmp = icmp ult i64 %i.0, 3
+  br i1 %cmp, label %for.cond.cleanup3, label %for.cond.cleanup
+
+for.cond.cleanup:                                 ; preds = %for.cond
+  %.lcssa10.lcssa = phi i16 [ %.lcssa10, %for.cond ]
+  %tmp3 = call i16 (i16) @func(i16 %.lcssa10.lcssa)
+  ret i16 0
+
+for.cond.cleanup3:                                ; preds = %for.cond
+  %.lcssa = phi i16 [ %tmp2, %for.cond ]
+  %inc9 = add i64 %i.0, 1
+  br label %for.cond
+}
+
+define i16 @partial_unroll(i16* %A) {
+; CHECK-LABEL: @partial_unroll(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br label [[FOR_COND:%.*]]
+; CHECK:       for.cond:
+; CHECK-NEXT:    [[I_0:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[INC9_2:%.*]], [[FOR_COND_CLEANUP3_2:%.*]] ]
+; CHECK-NEXT:    [[PTR:%.*]] = getelementptr inbounds i16, i16* [[A:%.*]], i64 [[I_0]]
+; CHECK-NEXT:    [[TMP2:%.*]] = load i16, i16* [[PTR]]
+; CHECK-NEXT:    br label [[FOR_COND_CLEANUP3:%.*]]
+; CHECK:       for.cond.cleanup:
+; CHECK-NEXT:    [[DOTLCSSA10_LCSSA:%.*]] = phi i16 [ [[TMP2_1:%.*]], [[FOR_COND_CLEANUP3_1:%.*]] ]
+; CHECK-NEXT:    [[TMP3:%.*]] = call i16 @func(i16 [[DOTLCSSA10_LCSSA]])
+; CHECK-NEXT:    ret i16 0
+; CHECK:       for.cond.cleanup3:
+; CHECK-NEXT:    [[INC9:%.*]] = add nuw nsw i64 [[I_0]], 1
+; CHECK-NEXT:    [[PTR_1:%.*]] = getelementptr inbounds i16, i16* [[A]], i64 [[INC9]]
+; CHECK-NEXT:    [[TMP2_1]] = load i16, i16* [[PTR_1]]
+; CHECK-NEXT:    br label [[FOR_COND_CLEANUP3_1]]
+; CHECK:       for.cond.cleanup3.1:
+; CHECK-NEXT:    [[INC9_1:%.*]] = add nuw nsw i64 [[INC9]], 1
+; CHECK-NEXT:    [[PTR_2:%.*]] = getelementptr inbounds i16, i16* [[A]], i64 [[INC9_1]]
+; CHECK-NEXT:    [[TMP2_2:%.*]] = load i16, i16* [[PTR_2]]
+; CHECK-NEXT:    [[CMP_2:%.*]] = icmp ult i64 [[INC9_1]], 200
+; CHECK-NEXT:    br i1 [[CMP_2]], label [[FOR_COND_CLEANUP3_2]], label [[FOR_COND_CLEANUP:%.*]]
+; CHECK:       for.cond.cleanup3.2:
+; CHECK-NEXT:    [[INC9_2]] = add nuw nsw i64 [[INC9_1]], 1
+; CHECK-NEXT:    br label [[FOR_COND]]
+;
+entry:
+  br label %for.cond
+
+for.cond:                                         ; preds = %for.cond.cleanup3, %entry
+  %.lcssa10 = phi i16 [ 123, %entry ], [ %.lcssa, %for.cond.cleanup3 ]
+  %i.0 = phi i64 [ 0, %entry ], [ %inc9, %for.cond.cleanup3 ]
+  %ptr = getelementptr inbounds i16, i16* %A, i64 %i.0
+  %tmp2 = load i16, i16* %ptr
+  %cmp = icmp ult i64 %i.0, 200
+  br i1 %cmp, label %for.cond.cleanup3, label %for.cond.cleanup
+
+for.cond.cleanup:                                 ; preds = %for.cond
+  %.lcssa10.lcssa = phi i16 [ %.lcssa10, %for.cond ]
+  %tmp3 = call i16 (i16) @func(i16 %.lcssa10.lcssa)
+  ret i16 0
+
+for.cond.cleanup3:                                ; preds = %for.cond
+  %.lcssa = phi i16 [ %tmp2, %for.cond ]
+  %inc9 = add i64 %i.0, 1
+  br label %for.cond
+}
+
+declare i16 @func(i16)
+