diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
index f2f8a85b7cc23..f68df8a9df86a 100644
--- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -8343,17 +8343,22 @@ VPRecipeBuilder::tryToWidenMemory(Instruction *I, ArrayRef<VPValue *> Operands,
     auto *GEP = dyn_cast<GetElementPtrInst>(
         Ptr->getUnderlyingValue()->stripPointerCasts());
     VPSingleDefRecipe *VectorPtr;
-    if (Reverse)
+    if (Reverse) {
+      // When folding the tail, we may compute an address that we don't in the
+      // original scalar loop and it may not be inbounds. Drop Inbounds in that
+      // case.
+      GEPNoWrapFlags Flags =
+          (CM.foldTailByMasking() || !GEP || !GEP->isInBounds())
+              ? GEPNoWrapFlags::none()
+              : GEPNoWrapFlags::inBounds();
       VectorPtr = new VPReverseVectorPointerRecipe(
-          Ptr, &Plan.getVF(), getLoadStoreType(I),
-          GEP && GEP->isInBounds() ? GEPNoWrapFlags::inBounds()
-                                   : GEPNoWrapFlags::none(),
-          I->getDebugLoc());
-    else
+          Ptr, &Plan.getVF(), getLoadStoreType(I), Flags, I->getDebugLoc());
+    } else {
       VectorPtr = new VPVectorPointerRecipe(Ptr, getLoadStoreType(I),
                                             GEP ? GEP->getNoWrapFlags()
                                                 : GEPNoWrapFlags::none(),
                                             I->getDebugLoc());
+    }
     Builder.getInsertBlock()->appendRecipe(VectorPtr);
     Ptr = VectorPtr;
   }
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-reverse-load-store.ll b/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-reverse-load-store.ll
index a330b6964a660..f323231445aad 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-reverse-load-store.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-reverse-load-store.ll
@@ -37,16 +37,16 @@ define void @reverse_load_store(i64 %startval, ptr noalias %ptr, ptr noalias %pt
 ; IF-EVL-NEXT:    [[TMP18:%.*]] = zext i32 [[TMP5]] to i64
 ; IF-EVL-NEXT:    [[TMP9:%.*]] = mul i64 0, [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP10:%.*]] = sub i64 1, [[TMP18]]
-; IF-EVL-NEXT:    [[TMP11:%.*]] = getelementptr inbounds i32, ptr [[TMP8]], i64 [[TMP9]]
-; IF-EVL-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i32, ptr [[TMP11]], i64 [[TMP10]]
+; IF-EVL-NEXT:    [[TMP16:%.*]] = getelementptr i32, ptr [[TMP8]], i64 [[TMP9]]
+; IF-EVL-NEXT:    [[TMP12:%.*]] = getelementptr i32, ptr [[TMP16]], i64 [[TMP10]]
 ; IF-EVL-NEXT:    [[VP_OP_LOAD:%.*]] = call <vscale x 4 x i32> @llvm.vp.load.nxv4i32.p0(ptr align 4 [[TMP12]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP5]])
 ; IF-EVL-NEXT:    [[VP_REVERSE:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vp.reverse.nxv4i32(<vscale x 4 x i32> [[VP_OP_LOAD]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP5]])
 ; IF-EVL-NEXT:    [[TMP13:%.*]] = getelementptr inbounds i32, ptr [[PTR2:%.*]], i64 [[TMP7]]
 ; IF-EVL-NEXT:    [[TMP19:%.*]] = zext i32 [[TMP5]] to i64
 ; IF-EVL-NEXT:    [[TMP14:%.*]] = mul i64 0, [[TMP19]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = sub i64 1, [[TMP19]]
-; IF-EVL-NEXT:    [[TMP16:%.*]] = getelementptr inbounds i32, ptr [[TMP13]], i64 [[TMP14]]
-; IF-EVL-NEXT:    [[TMP17:%.*]] = getelementptr inbounds i32, ptr [[TMP16]], i64 [[TMP15]]
+; IF-EVL-NEXT:    [[TMP22:%.*]] = getelementptr i32, ptr [[TMP13]], i64 [[TMP14]]
+; IF-EVL-NEXT:    [[TMP17:%.*]] = getelementptr i32, ptr [[TMP22]], i64 [[TMP15]]
 ; IF-EVL-NEXT:    [[VP_REVERSE3:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vp.reverse.nxv4i32(<vscale x 4 x i32> [[VP_REVERSE]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP5]])
 ; IF-EVL-NEXT:    call void @llvm.vp.store.nxv4i32.p0(<vscale x 4 x i32> [[VP_REVERSE3]], ptr align 4 [[TMP17]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP5]])
 ; IF-EVL-NEXT:    [[TMP20:%.*]] = zext i32 [[TMP5]] to i64
diff --git a/llvm/test/Transforms/LoopVectorize/X86/drop-inbounds-flags-for-reverse-vector-pointer.ll b/llvm/test/Transforms/LoopVectorize/X86/drop-inbounds-flags-for-reverse-vector-pointer.ll
index 66bb9357750c8..3d23090dd1235 100644
--- a/llvm/test/Transforms/LoopVectorize/X86/drop-inbounds-flags-for-reverse-vector-pointer.ll
+++ b/llvm/test/Transforms/LoopVectorize/X86/drop-inbounds-flags-for-reverse-vector-pointer.ll
@@ -4,8 +4,6 @@
 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"
 target triple = "x86_64-unknown-linux-gnu"
 
-; FIXME:  GEP flags on GEPs for reverse vector pointer need to be dropped when folding the tail.
-
 define i1 @fn(ptr %nno) #0 {
 ; CHECK-LABEL: define i1 @fn(
 ; CHECK-SAME: ptr [[NNO:%.*]]) #[[ATTR0:[0-9]+]] {
@@ -26,8 +24,8 @@ define i1 @fn(ptr %nno) #0 {
 ; CHECK-NEXT:    [[TMP2:%.*]] = and <4 x i64> [[VEC_IND]], splat (i64 1)
 ; CHECK-NEXT:    [[TMP3:%.*]] = icmp eq <4 x i64> [[TMP2]], zeroinitializer
 ; CHECK-NEXT:    [[TMP23:%.*]] = getelementptr inbounds nuw i32, ptr [[NNO]], i64 [[TMP22]]
-; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i32, ptr [[TMP23]], i32 0
-; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i32, ptr [[TMP5]], i32 -3
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr i32, ptr [[TMP23]], i32 0
+; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr i32, ptr [[TMP5]], i32 -3
 ; CHECK-NEXT:    [[REVERSE:%.*]] = shufflevector <4 x i1> [[TMP1]], <4 x i1> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
 ; CHECK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0(ptr [[TMP6]], i32 4, <4 x i1> [[REVERSE]], <4 x i32> poison)
 ; CHECK-NEXT:    [[REVERSE1:%.*]] = shufflevector <4 x i32> [[WIDE_MASKED_LOAD]], <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>