diff --git a/llvm/lib/Transforms/Scalar/IndVarSimplify.cpp b/llvm/lib/Transforms/Scalar/IndVarSimplify.cpp
index c32731185afd0..d434d26439c12 100644
--- a/llvm/lib/Transforms/Scalar/IndVarSimplify.cpp
+++ b/llvm/lib/Transforms/Scalar/IndVarSimplify.cpp
@@ -159,6 +159,8 @@ class IndVarSimplify {
 
   bool sinkUnusedInvariants(Loop *L);
 
+  bool isIVInitValTargetType(Value *InitVal, unsigned TargetTypeSize);
+
 public:
   IndVarSimplify(LoopInfo *LI, ScalarEvolution *SE, DominatorTree *DT,
                  const DataLayout &DL, TargetLibraryInfo *TLI,
@@ -816,6 +818,37 @@ static bool isLoopCounter(PHINode* Phi, Loop *L,
           isa<SCEVAddRecExpr>(SE->getSCEV(IncV)));
 }
 
+/// Determine if the initial value of IV is within the range of TargetTypeSize.
+/// If it is a constant, check if it falls within the representable range. If
+/// not, verify whether the width of the initial type exceeds TargetTypeSize.
+bool IndVarSimplify::isIVInitValTargetType(Value *InitVal,
+                                           unsigned TargetTypeSize) {
+
+  unsigned InitTypeSize = SE->getTypeSizeInBits(InitVal->getType());
+  if (InitTypeSize <= TargetTypeSize)
+    return true;
+
+  // If the initial value of the IV is a constant, check whether it is within
+  // the range.
+  if (auto *CI = dyn_cast<ConstantInt>(InitVal)) {
+
+    APInt Value = CI->getValue();
+    APInt MinRange = APInt(TargetTypeSize, 1ULL << (TargetTypeSize - 1));
+    APInt MaxRange = APInt(TargetTypeSize, (1ULL << (TargetTypeSize - 1)) - 1);
+    // Check if the constant is within a valid range
+    return (Value.sle(MaxRange) && Value.sge(-MinRange));
+  }
+
+  // Check the initial type of the initial value.
+  if (auto *SextInst = dyn_cast<SExtInst>(InitVal)) {
+    return isIVInitValTargetType(SextInst->getOperand(0), TargetTypeSize);
+  }
+  if (auto *ZextInst = dyn_cast<ZExtInst>(InitVal)) {
+    return isIVInitValTargetType(ZextInst->getOperand(0), TargetTypeSize);
+  }
+  return false;
+}
+
 /// Search the loop header for a loop counter (anadd rec w/step of one)
 /// suitable for use by LFTR.  If multiple counters are available, select the
 /// "best" one based profitable heuristics.
@@ -1048,9 +1081,27 @@ linearFunctionTestReplace(Loop *L, BasicBlock *ExitingBB,
     if (Extended) {
       bool Discard;
       L->makeLoopInvariant(ExitCnt, Discard);
-    } else
-      CmpIndVar = Builder.CreateTrunc(CmpIndVar, ExitCnt->getType(),
-                                      "lftr.wideiv");
+    } else {
+      // The design of LFTR requires:
+      // 1. The IV to be a LoopCounter, ensuring a step of 1.
+      // 2. The ICmpInst::Predicate can only be eq or ne, meaning that
+      //    ExitCnt must represent the final value of IV.
+      // Given these conditions, if the initial value 'start' of IV does not
+      // exceed ExitCntSize, then the range [start, end) of IV will stay
+      // within ExitCntSize. Consequently, the truncation will not result in
+      // signed or unsigned overflow.
+      // By legally adding the nsw/nuw flag to the Trunc instruction, avoid the
+      // unnecessary masking issue.
+      Value *InitVal = IndVar->getIncomingValueForBlock(L->getLoopPreheader());
+      if (isIVInitValTargetType(InitVal, ExitCntSize))
+        CmpIndVar = Builder.CreateTrunc(
+            CmpIndVar, ExitCnt->getType(), "lftr.wideiv",
+            cast<ICmpInst>(BI->getCondition())->isUnsigned(),
+            cast<ICmpInst>(BI->getCondition())->isSigned());
+      else
+        CmpIndVar =
+            Builder.CreateTrunc(CmpIndVar, ExitCnt->getType(), "lftr.wideiv");
+    }
   }
   LLVM_DEBUG(dbgs() << "INDVARS: Rewriting loop exit condition to:\n"
                     << "      LHS:" << *CmpIndVar << '\n'
diff --git a/llvm/test/Transforms/IndVarSimplify/lftr-masking.ll b/llvm/test/Transforms/IndVarSimplify/lftr-masking.ll
new file mode 100644
index 0000000000000..ef0c0916f2dd5
--- /dev/null
+++ b/llvm/test/Transforms/IndVarSimplify/lftr-masking.ll
@@ -0,0 +1,45 @@
+; Legally add nsw/nuw flag for the Trunc instruction.
+; RUN: opt < %s -passes='indvars' -S  | FileCheck %s
+
+target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128"
+
+define dso_local void @func(ptr noundef captures(none) %0, i32 noundef %1) local_unnamed_addr {
+; CHECK-LABEL: define dso_local void @func(ptr noundef captures(none) %0, i32 noundef %1) local_unnamed_addr {
+; CHECK-NEXT:  [[CHECK:%.*]] = icmp slt i32 %1, 100
+; CHECK-NEXT:  br i1 [[CHECK]], label [[LOOP_PREHEADER:%.*]], label [[LOOP_EXIT:%.*]]
+; CHECK:       loop.preheader:
+; CHECK-NEXT:  [[SEXT_START:%.*]] = sext i32 %1 to i64
+; CHECK-NEXT:  br label [[LOOP_BODY:%.*]]
+; CHECK:       loop.exit.loopexit:
+; CHECK-NEXT:  br label [[LOOPEXIT:%.*]]
+; CHECK:       loop.exit:
+; CHECK-NEXT:  ret void
+; CHECK:       loop.body:
+; CHECK-NEXT:  [[IV:%.*]] = phi i64 [ [[POSTINC:%.*]], [[LOOP_BODY]] ], [ [[SEXT_START:%.*]], [[LOOP_PREHEADER]] ]
+; CHECK-NEXT:  [[GEP:%.*]] = getelementptr inbounds i32, ptr %0, i64 [[IV]]
+; CHECK-NEXT:  [[LOAD:%.*]] = load i32, ptr [[GEP]], align 4
+; CHECK-NEXT:  [[DATA:%.*]] = add nsw i32 [[LOAD]], 1
+; CHECK-NEXT:  store i32 [[DATA]], ptr [[GEP]], align 4
+; CHECK-NEXT:  [[POSTINC:%.*]] = add nsw i64 [[IV]], 1
+; CHECK-NEXT:  [[LFTR_WIDEIV:%.*]] = trunc nsw i64 [[POSTINC]] to i32
+; CHECK-NEXT:  [[EXITCOND:%.*]] = icmp ne i32 [[LFTR_WIDEIV]], 100
+; CHECK-NEXT:  br i1 [[EXITCOND]], label [[LOOP_BODY]], label [[LOOP_EXIT_LOOPEXIT:%.*]]
+; CHECK-NEXT:  }
+  %3 = icmp slt i32 %1, 100
+  br i1 %3, label %loop.preheader, label %loop.exit
+loop.preheader:                                                
+  br label %loop.body
+
+loop.exit: 
+  ret void
+loop.body:                                                
+  %8 = phi i32 [ %13, %loop.body ], [ %1, %loop.preheader ]
+  %9 = sext i32 %8 to i64
+  %10 = getelementptr inbounds i32, ptr %0, i64 %9
+  %11 = load i32, ptr %10, align 4
+  %12 = add nsw i32 %11, 1
+  store i32 %12, ptr %10, align 4
+  %13 = add nsw i32 %8, 1
+  %14 = icmp slt i32 %8, 99
+  br i1 %14, label %loop.body, label %loop.exit
+}
diff --git a/llvm/test/Transforms/PhaseOrdering/AArch64/constraint-elimination-placement.ll b/llvm/test/Transforms/PhaseOrdering/AArch64/constraint-elimination-placement.ll
index bbdbd95c6017a..ddd98f21c36a4 100644
--- a/llvm/test/Transforms/PhaseOrdering/AArch64/constraint-elimination-placement.ll
+++ b/llvm/test/Transforms/PhaseOrdering/AArch64/constraint-elimination-placement.ll
@@ -33,8 +33,7 @@ define i1 @test_order_1(ptr %this, ptr noalias %other, i1 %tobool9.not, i32 %cal
 ; CHECK-NEXT:    br i1 [[CMP44]], label [[FOR_BODY45]], label [[FOR_COND]]
 ; CHECK:       for.inc57:
 ; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add nsw i64 [[INDVARS_IV]], 1
-; CHECK-NEXT:    [[TMP1:%.*]] = and i64 [[INDVARS_IV_NEXT]], 4294967295
-; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp eq i64 [[TMP1]], 1
+; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp eq i64 [[INDVARS_IV]], 0
 ; CHECK-NEXT:    br i1 [[EXITCOND]], label [[FOR_COND41_PREHEADER_PREHEADER]], label [[FOR_COND41_PREHEADER]]
 ; CHECK:       exit:
 ; CHECK-NEXT:    ret i1 false