Fix a bug in LinearFunctionTestReplace that created invalid loop exit checks.

Reviewed by Andy

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193303 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Transforms/Scalar/IndVarSimplify.cpp b/lib/Transforms/Scalar/IndVarSimplify.cpp
index 7785855d0e13c00f08ac91bd2ab5416f619c080f..cfd8db0f6cc4097c9c3d129249b66f7ed8f3daf3 100644 (file)
--- a/lib/Transforms/Scalar/IndVarSimplify.cpp
+++ b/lib/Transforms/Scalar/IndVarSimplify.cpp
@@ -1479,8 +1479,14 @@ static Value *genLoopLimit(PHINode *IndVar, const SCEV *IVCount, Loop *L,
   if (IndVar->getType()->isPointerTy()
       && !IVCount->getType()->isPointerTy()) {
 
+    // IVOffset will be the new GEP offset that is interpreted by GEP as a
+    // signed value. IVCount on the other hand represents the loop trip count,
+    // which is an unsigned value. FindLoopCounter only allows induction
+    // variables that have a positive unit stride of one. This means we don't
+    // have to handle the case of negative offsets (yet) and just need to zero
+    // extend IVCount.
     Type *OfsTy = SE->getEffectiveSCEVType(IVInit->getType());
-    const SCEV *IVOffset = SE->getTruncateOrSignExtend(IVCount, OfsTy);
+    const SCEV *IVOffset = SE->getTruncateOrZeroExtend(IVCount, OfsTy);
 
     // Expand the code for the iteration count.
     assert(SE->isLoopInvariant(IVOffset, L) &&

diff --git a/test/Transforms/IndVarSimplify/lftr-zext.ll b/test/Transforms/IndVarSimplify/lftr-zext.ll
new file mode 100644 (file)
index 0000000..32fa61a
--- /dev/null
+++ b/test/Transforms/IndVarSimplify/lftr-zext.ll
@@ -0,0 +1,26 @@
+; RUN: opt < %s -indvars -S | FileCheck %s
+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-apple-macosx10.9.0"
+
+@data = common global [240 x i8] zeroinitializer, align 16
+
+define void @foo(i8* %a) nounwind uwtable ssp {
+; CHECK: %exitcond
+; CHECK-NOT: ([240 x i8]* @data, i64 0, i64 -16)
+  br label %1
+
+; <label>:1                                       ; preds = %0, %1
+  %i.0 = phi i8 [ 0, %0 ], [ %5, %1 ]
+  %p.0 = phi i8* [ getelementptr inbounds ([240 x i8]* @data, i64 0, i64 0), %0 ], [ %4, %1 ]
+  %.0 = phi i8* [ %a, %0 ], [ %2, %1 ]
+  %2 = getelementptr inbounds i8* %.0, i64 1
+  %3 = load i8* %.0, align 1
+  %4 = getelementptr inbounds i8* %p.0, i64 1
+  store i8 %3, i8* %p.0, align 1
+  %5 = add i8 %i.0, 1
+  %6 = icmp ult i8 %5, -16
+  br i1 %6, label %1, label %7
+
+; <label>:7                                       ; preds = %1
+  ret void
+}