LoopVectorizer: Fix a bug in the code that updates the loop exiting block.

LCSSA PHIs may have undef values. The vectorizer updates values that are used by outside users such as PHIs.
The bug happened because undefs are not loop values. This patch handles these PHIs.

PR14725

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171251 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Transforms/Vectorize/LoopVectorize.cpp b/lib/Transforms/Vectorize/LoopVectorize.cpp
index 7fb9bbada0f7f7c16570b83ef879ac58c22c6f53..653c111182c66062f7b7a97e8ae1b5959f4830c5 100644 (file)
--- a/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -909,6 +909,19 @@ InnerLoopVectorizer::vectorizeLoop(LoopVectorizationLegality *Legal) {
     (RdxPhi)->setIncomingValue(SelfEdgeBlockIdx, Scalar0);
     (RdxPhi)->setIncomingValue(IncomingEdgeBlockIdx, RdxDesc.LoopExitInstr);
   }// end of for each redux variable.
+
+  // The Loop exit block may have single value PHI nodes where the incoming
+  // value is 'undef'. While vectorizing we only handled real values that
+  // were defined inside the loop. Here we handle the 'undef case'.
+  // See PR14725.
+  for (BasicBlock::iterator LEI = LoopExitBlock->begin(),
+       LEE = LoopExitBlock->end(); LEI != LEE; ++LEI) {
+    PHINode *LCSSAPhi = dyn_cast<PHINode>(LEI);
+    if (!LCSSAPhi) continue;
+    if (LCSSAPhi->getNumIncomingValues() == 1)
+      LCSSAPhi->addIncoming(UndefValue::get(LCSSAPhi->getType()),
+                            LoopMiddleBlock);
+  }
 }
 
 Value *InnerLoopVectorizer::createEdgeMask(BasicBlock *Src, BasicBlock *Dst) {

diff --git a/test/Transforms/LoopVectorize/lcssa-crash.ll b/test/Transforms/LoopVectorize/lcssa-crash.ll
new file mode 100644 (file)
index 0000000..a618db5
--- /dev/null
+++ b/test/Transforms/LoopVectorize/lcssa-crash.ll
@@ -0,0 +1,29 @@
+; RUN: opt < %s  -loop-vectorize -force-vector-width=4 
+
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+%type1 = type { %type2 }
+%type2 = type { [0 x i8*], i8**, i32, i32, i32 }
+
+define void @test() nounwind uwtable align 2 {
+  br label %for.body.lr.ph.i.i.i
+
+for.body.lr.ph.i.i.i:
+  br label %for.body.i.i.i
+
+for.body.i.i.i:
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc.i.i.i ], [ 0, %for.body.lr.ph.i.i.i ]
+  br label %for.inc.i.i.i
+
+for.inc.i.i.i:
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp ne i32 %lftr.wideiv, undef
+  br i1 %exitcond, label %for.body.i.i.i, label %for.end.i.i.i
+
+for.end.i.i.i:
+  %lcssa = phi %type1* [ undef, %for.inc.i.i.i ]
+  unreachable
+}
+