[RISCV][TTI] Enable masked interleave access for scalable vector (#149981)

Now that support for masked loads/stores of interleave groups has
landed, we can enable the loop vectorizer to generate masked interleave
access where applicable.

This improves vectorization in several ways:
* Internal predication support: This enables interleave group
vectorization for loops with internal control flow predication, provided
all members of the group share the same predicate. Gaps in interleave
groups are still not efficiently handled by masking, so masking for gaps
remains disabled for now.
* Tail folding: This allows tail folding of loops with interleave groups
by using masking. Without this, vectorized loops with interleaves would
fall back to using separate gather/scatter accesses, which can be
significantly less efficient.
* Scalable vector support: Currently, only scalable vector types are
supported for masked interleave lowering. Fixed-length vector support
will be enabled in the future.

As interleave access is not yet supported with tail folding by EVL, that
functionality is temporarily disabled. We are going to create another
patch to support it.

Co-authored-by: Philip Reames <preames@rivosinc.com>

---------

Co-authored-by: Philip Reames <preames@rivosinc.com>

5 files changed, 152 insertions, 193 deletions

diff --git a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
index fd634b5..1624f12a 100644
--- a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
+++ b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
@@ -979,10 +979,12 @@ InstructionCost RISCVTTIImpl::getInterleavedMemoryOpCost(
     Align Alignment, unsigned AddressSpace, TTI::TargetCostKind CostKind,
     bool UseMaskForCond, bool UseMaskForGaps) const {
 
-  // The interleaved memory access pass will lower interleaved memory ops (i.e
-  // a load and store followed by a specific shuffle) to vlseg/vsseg
-  // intrinsics.
-  if (!UseMaskForCond && !UseMaskForGaps &&
+  // The interleaved memory access pass will lower (de)interleave ops combined
+  // with an adjacent appropriate memory to vlseg/vsseg intrinsics. vlseg/vsseg
+  // only support masking per-iteration (i.e. condition), not per-segment (i.e.
+  // gap).
+  // TODO: Support masked interleaved access for fixed length vector.
+  if ((isa<ScalableVectorType>(VecTy) || !UseMaskForCond) && !UseMaskForGaps &&
       Factor <= TLI->getMaxSupportedInterleaveFactor()) {
     auto *VTy = cast<VectorType>(VecTy);
     std::pair<InstructionCost, MVT> LT = getTypeLegalizationCost(VTy);
diff --git a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.h b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.h
index d62d99c..05d504c 100644
--- a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.h
+++ b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.h
@@ -398,6 +398,10 @@ public:
 
   bool enableInterleavedAccessVectorization() const override { return true; }
 
+  bool enableMaskedInterleavedAccessVectorization() const override {
+    return ST->hasVInstructions();
+  }
+
   unsigned getMinTripCountTailFoldingThreshold() const override;
 
   enum RISCVRegisterClass { GPRRC, FPRRC, VRRC };
diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
index 99a96a8..9c9201c 100644
--- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -1359,7 +1359,9 @@ public:
       return;
     // Override EVL styles if needed.
     // FIXME: Investigate opportunity for fixed vector factor.
+    // FIXME: Support interleave accesses.
     bool EVLIsLegal = UserIC <= 1 && IsScalableVF &&
+                      !InterleaveInfo.hasGroups() &&
                       TTI.hasActiveVectorLength() && !EnableVPlanNativePath;
     if (EVLIsLegal)
       return;
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/interleaved-masked-access.ll b/llvm/test/Transforms/LoopVectorize/RISCV/interleaved-masked-access.ll
index dbe6f27..2a4cd28 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/interleaved-masked-access.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/interleaved-masked-access.ll
@@ -30,26 +30,25 @@ define void @masked_strided_factor2(ptr noalias nocapture readonly %p, ptr noali
 ; SCALAR_EPILOGUE-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; SCALAR_EPILOGUE-NEXT:    [[VEC_IND:%.*]] = phi <vscale x 16 x i32> [ [[TMP5]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; SCALAR_EPILOGUE-NEXT:    [[TMP6:%.*]] = icmp ugt <vscale x 16 x i32> [[VEC_IND]], [[BROADCAST_SPLAT]]
-; SCALAR_EPILOGUE-NEXT:    [[TMP7:%.*]] = shl nuw nsw <vscale x 16 x i32> [[VEC_IND]], splat (i32 1)
-; SCALAR_EPILOGUE-NEXT:    [[TMP8:%.*]] = zext nneg <vscale x 16 x i32> [[TMP7]] to <vscale x 16 x i64>
-; SCALAR_EPILOGUE-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i8, ptr [[P]], <vscale x 16 x i64> [[TMP8]]
-; SCALAR_EPILOGUE-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 16 x i8> @llvm.masked.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> [[TMP9]], i32 1, <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i8> poison)
-; SCALAR_EPILOGUE-NEXT:    [[TMP10:%.*]] = or disjoint <vscale x 16 x i32> [[TMP7]], splat (i32 1)
-; SCALAR_EPILOGUE-NEXT:    [[TMP11:%.*]] = zext nneg <vscale x 16 x i32> [[TMP10]] to <vscale x 16 x i64>
-; SCALAR_EPILOGUE-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i8, ptr [[P]], <vscale x 16 x i64> [[TMP11]]
-; SCALAR_EPILOGUE-NEXT:    [[WIDE_MASKED_GATHER3:%.*]] = call <vscale x 16 x i8> @llvm.masked.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> [[TMP12]], i32 1, <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i8> poison)
-; SCALAR_EPILOGUE-NEXT:    [[TMP13:%.*]] = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> [[WIDE_MASKED_GATHER]], <vscale x 16 x i8> [[WIDE_MASKED_GATHER3]])
-; SCALAR_EPILOGUE-NEXT:    [[TMP14:%.*]] = zext nneg <vscale x 16 x i32> [[TMP7]] to <vscale x 16 x i64>
-; SCALAR_EPILOGUE-NEXT:    [[TMP15:%.*]] = getelementptr inbounds i8, ptr [[Q]], <vscale x 16 x i64> [[TMP14]]
-; SCALAR_EPILOGUE-NEXT:    call void @llvm.masked.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> [[TMP13]], <vscale x 16 x ptr> [[TMP15]], i32 1, <vscale x 16 x i1> [[TMP6]])
-; SCALAR_EPILOGUE-NEXT:    [[TMP16:%.*]] = sub <vscale x 16 x i8> zeroinitializer, [[TMP13]]
-; SCALAR_EPILOGUE-NEXT:    [[TMP17:%.*]] = zext nneg <vscale x 16 x i32> [[TMP10]] to <vscale x 16 x i64>
-; SCALAR_EPILOGUE-NEXT:    [[TMP18:%.*]] = getelementptr inbounds i8, ptr [[Q]], <vscale x 16 x i64> [[TMP17]]
-; SCALAR_EPILOGUE-NEXT:    call void @llvm.masked.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> [[TMP16]], <vscale x 16 x ptr> [[TMP18]], i32 1, <vscale x 16 x i1> [[TMP6]])
+; SCALAR_EPILOGUE-NEXT:    [[TMP7:%.*]] = shl i32 [[INDEX]], 1
+; SCALAR_EPILOGUE-NEXT:    [[TMP8:%.*]] = sext i32 [[TMP7]] to i64
+; SCALAR_EPILOGUE-NEXT:    [[TMP9:%.*]] = getelementptr i8, ptr [[P]], i64 [[TMP8]]
+; SCALAR_EPILOGUE-NEXT:    [[INTERLEAVED_MASK:%.*]] = call <vscale x 32 x i1> @llvm.vector.interleave2.nxv32i1(<vscale x 16 x i1> [[TMP6]], <vscale x 16 x i1> [[TMP6]])
+; SCALAR_EPILOGUE-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 32 x i8> @llvm.masked.load.nxv32i8.p0(ptr [[TMP9]], i32 1, <vscale x 32 x i1> [[INTERLEAVED_MASK]], <vscale x 32 x i8> poison)
+; SCALAR_EPILOGUE-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.vector.deinterleave2.nxv32i8(<vscale x 32 x i8> [[WIDE_MASKED_VEC]])
+; SCALAR_EPILOGUE-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 0
+; SCALAR_EPILOGUE-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 1
+; SCALAR_EPILOGUE-NEXT:    [[TMP12:%.*]] = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> [[TMP10]], <vscale x 16 x i8> [[TMP11]])
+; SCALAR_EPILOGUE-NEXT:    [[TMP13:%.*]] = sext i32 [[TMP7]] to i64
+; SCALAR_EPILOGUE-NEXT:    [[TMP14:%.*]] = getelementptr i8, ptr [[Q]], i64 [[TMP13]]
+; SCALAR_EPILOGUE-NEXT:    [[TMP15:%.*]] = sub <vscale x 16 x i8> zeroinitializer, [[TMP12]]
+; SCALAR_EPILOGUE-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 32 x i8> @llvm.vector.interleave2.nxv32i8(<vscale x 16 x i8> [[TMP12]], <vscale x 16 x i8> [[TMP15]])
+; SCALAR_EPILOGUE-NEXT:    [[INTERLEAVED_MASK3:%.*]] = call <vscale x 32 x i1> @llvm.vector.interleave2.nxv32i1(<vscale x 16 x i1> [[TMP6]], <vscale x 16 x i1> [[TMP6]])
+; SCALAR_EPILOGUE-NEXT:    call void @llvm.masked.store.nxv32i8.p0(<vscale x 32 x i8> [[INTERLEAVED_VEC]], ptr [[TMP14]], i32 1, <vscale x 32 x i1> [[INTERLEAVED_MASK3]])
 ; SCALAR_EPILOGUE-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[INDEX]], [[TMP4]]
 ; SCALAR_EPILOGUE-NEXT:    [[VEC_IND_NEXT]] = add <vscale x 16 x i32> [[VEC_IND]], [[BROADCAST_SPLAT2]]
-; SCALAR_EPILOGUE-NEXT:    [[TMP19:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
-; SCALAR_EPILOGUE-NEXT:    br i1 [[TMP19]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; SCALAR_EPILOGUE-NEXT:    [[TMP16:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
+; SCALAR_EPILOGUE-NEXT:    br i1 [[TMP16]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
 ; SCALAR_EPILOGUE:       middle.block:
 ; SCALAR_EPILOGUE-NEXT:    [[CMP_N:%.*]] = icmp eq i32 [[N_MOD_VF]], 0
 ; SCALAR_EPILOGUE-NEXT:    br i1 [[CMP_N]], label [[FOR_END:%.*]], label [[SCALAR_PH]]
@@ -80,26 +79,25 @@ define void @masked_strided_factor2(ptr noalias nocapture readonly %p, ptr noali
 ; PREDICATED_DATA-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <vscale x 16 x i1> @llvm.get.active.lane.mask.nxv16i1.i32(i32 [[INDEX]], i32 1024)
 ; PREDICATED_DATA-NEXT:    [[TMP5:%.*]] = icmp ugt <vscale x 16 x i32> [[VEC_IND]], [[BROADCAST_SPLAT]]
 ; PREDICATED_DATA-NEXT:    [[TMP6:%.*]] = select <vscale x 16 x i1> [[ACTIVE_LANE_MASK]], <vscale x 16 x i1> [[TMP5]], <vscale x 16 x i1> zeroinitializer
-; PREDICATED_DATA-NEXT:    [[TMP7:%.*]] = shl nuw nsw <vscale x 16 x i32> [[VEC_IND]], splat (i32 1)
-; PREDICATED_DATA-NEXT:    [[TMP8:%.*]] = zext nneg <vscale x 16 x i32> [[TMP7]] to <vscale x 16 x i64>
-; PREDICATED_DATA-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i8, ptr [[P]], <vscale x 16 x i64> [[TMP8]]
-; PREDICATED_DATA-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 16 x i8> @llvm.masked.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> [[TMP9]], i32 1, <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i8> poison)
-; PREDICATED_DATA-NEXT:    [[TMP10:%.*]] = or disjoint <vscale x 16 x i32> [[TMP7]], splat (i32 1)
-; PREDICATED_DATA-NEXT:    [[TMP11:%.*]] = zext nneg <vscale x 16 x i32> [[TMP10]] to <vscale x 16 x i64>
-; PREDICATED_DATA-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i8, ptr [[P]], <vscale x 16 x i64> [[TMP11]]
-; PREDICATED_DATA-NEXT:    [[WIDE_MASKED_GATHER3:%.*]] = call <vscale x 16 x i8> @llvm.masked.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> [[TMP12]], i32 1, <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i8> poison)
-; PREDICATED_DATA-NEXT:    [[TMP13:%.*]] = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> [[WIDE_MASKED_GATHER]], <vscale x 16 x i8> [[WIDE_MASKED_GATHER3]])
-; PREDICATED_DATA-NEXT:    [[TMP14:%.*]] = zext nneg <vscale x 16 x i32> [[TMP7]] to <vscale x 16 x i64>
-; PREDICATED_DATA-NEXT:    [[TMP15:%.*]] = getelementptr inbounds i8, ptr [[Q]], <vscale x 16 x i64> [[TMP14]]
-; PREDICATED_DATA-NEXT:    call void @llvm.masked.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> [[TMP13]], <vscale x 16 x ptr> [[TMP15]], i32 1, <vscale x 16 x i1> [[TMP6]])
-; PREDICATED_DATA-NEXT:    [[TMP16:%.*]] = sub <vscale x 16 x i8> zeroinitializer, [[TMP13]]
-; PREDICATED_DATA-NEXT:    [[TMP17:%.*]] = zext nneg <vscale x 16 x i32> [[TMP10]] to <vscale x 16 x i64>
-; PREDICATED_DATA-NEXT:    [[TMP18:%.*]] = getelementptr inbounds i8, ptr [[Q]], <vscale x 16 x i64> [[TMP17]]
-; PREDICATED_DATA-NEXT:    call void @llvm.masked.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> [[TMP16]], <vscale x 16 x ptr> [[TMP18]], i32 1, <vscale x 16 x i1> [[TMP6]])
+; PREDICATED_DATA-NEXT:    [[TMP7:%.*]] = shl i32 [[INDEX]], 1
+; PREDICATED_DATA-NEXT:    [[TMP8:%.*]] = sext i32 [[TMP7]] to i64
+; PREDICATED_DATA-NEXT:    [[TMP9:%.*]] = getelementptr i8, ptr [[P]], i64 [[TMP8]]
+; PREDICATED_DATA-NEXT:    [[INTERLEAVED_MASK:%.*]] = call <vscale x 32 x i1> @llvm.vector.interleave2.nxv32i1(<vscale x 16 x i1> [[TMP6]], <vscale x 16 x i1> [[TMP6]])
+; PREDICATED_DATA-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 32 x i8> @llvm.masked.load.nxv32i8.p0(ptr [[TMP9]], i32 1, <vscale x 32 x i1> [[INTERLEAVED_MASK]], <vscale x 32 x i8> poison)
+; PREDICATED_DATA-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.vector.deinterleave2.nxv32i8(<vscale x 32 x i8> [[WIDE_MASKED_VEC]])
+; PREDICATED_DATA-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 0
+; PREDICATED_DATA-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 1
+; PREDICATED_DATA-NEXT:    [[TMP12:%.*]] = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> [[TMP10]], <vscale x 16 x i8> [[TMP11]])
+; PREDICATED_DATA-NEXT:    [[TMP13:%.*]] = sext i32 [[TMP7]] to i64
+; PREDICATED_DATA-NEXT:    [[TMP14:%.*]] = getelementptr i8, ptr [[Q]], i64 [[TMP13]]
+; PREDICATED_DATA-NEXT:    [[TMP15:%.*]] = sub <vscale x 16 x i8> zeroinitializer, [[TMP12]]
+; PREDICATED_DATA-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 32 x i8> @llvm.vector.interleave2.nxv32i8(<vscale x 16 x i8> [[TMP12]], <vscale x 16 x i8> [[TMP15]])
+; PREDICATED_DATA-NEXT:    [[INTERLEAVED_MASK3:%.*]] = call <vscale x 32 x i1> @llvm.vector.interleave2.nxv32i1(<vscale x 16 x i1> [[TMP6]], <vscale x 16 x i1> [[TMP6]])
+; PREDICATED_DATA-NEXT:    call void @llvm.masked.store.nxv32i8.p0(<vscale x 32 x i8> [[INTERLEAVED_VEC]], ptr [[TMP14]], i32 1, <vscale x 32 x i1> [[INTERLEAVED_MASK3]])
 ; PREDICATED_DATA-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[INDEX]], [[TMP3]]
 ; PREDICATED_DATA-NEXT:    [[VEC_IND_NEXT]] = add <vscale x 16 x i32> [[VEC_IND]], [[BROADCAST_SPLAT2]]
-; PREDICATED_DATA-NEXT:    [[TMP19:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
-; PREDICATED_DATA-NEXT:    br i1 [[TMP19]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; PREDICATED_DATA-NEXT:    [[TMP16:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
+; PREDICATED_DATA-NEXT:    br i1 [[TMP16]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
 ; PREDICATED_DATA:       middle.block:
 ; PREDICATED_DATA-NEXT:    br label [[FOR_END:%.*]]
 ; PREDICATED_DATA:       scalar.ph:
@@ -120,37 +118,34 @@ define void @masked_strided_factor2(ptr noalias nocapture readonly %p, ptr noali
 ; PREDICATED_DATA-WITH-EVL-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 16 x i32> poison, i32 [[CONV]], i64 0
 ; PREDICATED_DATA-WITH-EVL-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 16 x i32> [[BROADCAST_SPLATINSERT]], <vscale x 16 x i32> poison, <vscale x 16 x i32> zeroinitializer
 ; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP4:%.*]] = call <vscale x 16 x i32> @llvm.stepvector.nxv16i32()
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <vscale x 16 x i32> poison, i32 [[TMP3]], i64 0
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[BROADCAST_SPLAT2:%.*]] = shufflevector <vscale x 16 x i32> [[BROADCAST_SPLATINSERT1]], <vscale x 16 x i32> poison, <vscale x 16 x i32> zeroinitializer
 ; PREDICATED_DATA-WITH-EVL-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; PREDICATED_DATA-WITH-EVL:       vector.body:
 ; PREDICATED_DATA-WITH-EVL-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[EVL_BASED_IV:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_EVL_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; PREDICATED_DATA-WITH-EVL-NEXT:    [[VEC_IND:%.*]] = phi <vscale x 16 x i32> [ [[TMP4]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[AVL:%.*]] = sub i32 1024, [[EVL_BASED_IV]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP5:%.*]] = call i32 @llvm.experimental.get.vector.length.i32(i32 [[AVL]], i32 16, i1 true)
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <vscale x 16 x i32> poison, i32 [[TMP5]], i64 0
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[BROADCAST_SPLAT2:%.*]] = shufflevector <vscale x 16 x i32> [[BROADCAST_SPLATINSERT1]], <vscale x 16 x i32> poison, <vscale x 16 x i32> zeroinitializer
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP5:%.*]] = icmp ult <vscale x 16 x i32> [[VEC_IND]], splat (i32 1024)
 ; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP6:%.*]] = icmp ugt <vscale x 16 x i32> [[VEC_IND]], [[BROADCAST_SPLAT]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP7:%.*]] = shl nuw nsw <vscale x 16 x i32> [[VEC_IND]], splat (i32 1)
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP8:%.*]] = zext nneg <vscale x 16 x i32> [[TMP7]] to <vscale x 16 x i64>
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i8, ptr [[P]], <vscale x 16 x i64> [[TMP8]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 16 x i8> @llvm.vp.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> align 1 [[TMP9]], <vscale x 16 x i1> [[TMP6]], i32 [[TMP5]])
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP10:%.*]] = or disjoint <vscale x 16 x i32> [[TMP7]], splat (i32 1)
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP11:%.*]] = zext nneg <vscale x 16 x i32> [[TMP10]] to <vscale x 16 x i64>
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i8, ptr [[P]], <vscale x 16 x i64> [[TMP11]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[WIDE_MASKED_GATHER3:%.*]] = call <vscale x 16 x i8> @llvm.vp.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> align 1 [[TMP12]], <vscale x 16 x i1> [[TMP6]], i32 [[TMP5]])
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP13:%.*]] = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> [[WIDE_MASKED_GATHER]], <vscale x 16 x i8> [[WIDE_MASKED_GATHER3]])
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP14:%.*]] = zext nneg <vscale x 16 x i32> [[TMP7]] to <vscale x 16 x i64>
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP15:%.*]] = getelementptr inbounds i8, ptr [[Q]], <vscale x 16 x i64> [[TMP14]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    call void @llvm.vp.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> [[TMP13]], <vscale x 16 x ptr> align 1 [[TMP15]], <vscale x 16 x i1> [[TMP6]], i32 [[TMP5]])
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP7:%.*]] = select <vscale x 16 x i1> [[TMP5]], <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i1> zeroinitializer
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP8:%.*]] = shl i32 [[INDEX]], 1
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP9:%.*]] = sext i32 [[TMP8]] to i64
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP10:%.*]] = getelementptr i8, ptr [[P]], i64 [[TMP9]]
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[INTERLEAVED_MASK:%.*]] = call <vscale x 32 x i1> @llvm.vector.interleave2.nxv32i1(<vscale x 16 x i1> [[TMP7]], <vscale x 16 x i1> [[TMP7]])
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 32 x i8> @llvm.masked.load.nxv32i8.p0(ptr [[TMP10]], i32 1, <vscale x 32 x i1> [[INTERLEAVED_MASK]], <vscale x 32 x i8> poison)
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.vector.deinterleave2.nxv32i8(<vscale x 32 x i8> [[WIDE_MASKED_VEC]])
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 0
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 1
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP13:%.*]] = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> [[TMP11]], <vscale x 16 x i8> [[TMP12]])
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP14:%.*]] = sext i32 [[TMP8]] to i64
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP15:%.*]] = getelementptr i8, ptr [[Q]], i64 [[TMP14]]
 ; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP16:%.*]] = sub <vscale x 16 x i8> zeroinitializer, [[TMP13]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP17:%.*]] = zext nneg <vscale x 16 x i32> [[TMP10]] to <vscale x 16 x i64>
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP18:%.*]] = getelementptr inbounds i8, ptr [[Q]], <vscale x 16 x i64> [[TMP17]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    call void @llvm.vp.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> [[TMP16]], <vscale x 16 x ptr> align 1 [[TMP18]], <vscale x 16 x i1> [[TMP6]], i32 [[TMP5]])
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i32 [[TMP5]], [[EVL_BASED_IV]]
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 32 x i8> @llvm.vector.interleave2.nxv32i8(<vscale x 16 x i8> [[TMP13]], <vscale x 16 x i8> [[TMP16]])
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[INTERLEAVED_MASK3:%.*]] = call <vscale x 32 x i1> @llvm.vector.interleave2.nxv32i1(<vscale x 16 x i1> [[TMP7]], <vscale x 16 x i1> [[TMP7]])
+; PREDICATED_DATA-WITH-EVL-NEXT:    call void @llvm.masked.store.nxv32i8.p0(<vscale x 32 x i8> [[INTERLEAVED_VEC]], ptr [[TMP15]], i32 1, <vscale x 32 x i1> [[INTERLEAVED_MASK3]])
 ; PREDICATED_DATA-WITH-EVL-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[INDEX]], [[TMP3]]
 ; PREDICATED_DATA-WITH-EVL-NEXT:    [[VEC_IND_NEXT]] = add <vscale x 16 x i32> [[VEC_IND]], [[BROADCAST_SPLAT2]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP19:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    br i1 [[TMP19]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP17:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
+; PREDICATED_DATA-WITH-EVL-NEXT:    br i1 [[TMP17]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
 ; PREDICATED_DATA-WITH-EVL:       middle.block:
 ; PREDICATED_DATA-WITH-EVL-NEXT:    br label [[FOR_END:%.*]]
 ; PREDICATED_DATA-WITH-EVL:       scalar.ph:
@@ -215,42 +210,29 @@ define void @masked_strided_factor4(ptr noalias nocapture readonly %p, ptr noali
 ; SCALAR_EPILOGUE-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; SCALAR_EPILOGUE-NEXT:    [[VEC_IND:%.*]] = phi <vscale x 16 x i32> [ [[TMP5]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; SCALAR_EPILOGUE-NEXT:    [[TMP6:%.*]] = icmp ugt <vscale x 16 x i32> [[VEC_IND]], [[BROADCAST_SPLAT]]
-; SCALAR_EPILOGUE-NEXT:    [[TMP7:%.*]] = shl nuw nsw <vscale x 16 x i32> [[VEC_IND]], splat (i32 2)
-; SCALAR_EPILOGUE-NEXT:    [[TMP8:%.*]] = or disjoint <vscale x 16 x i32> [[TMP7]], splat (i32 1)
-; SCALAR_EPILOGUE-NEXT:    [[TMP9:%.*]] = or disjoint <vscale x 16 x i32> [[TMP7]], splat (i32 2)
-; SCALAR_EPILOGUE-NEXT:    [[TMP10:%.*]] = or disjoint <vscale x 16 x i32> [[TMP7]], splat (i32 3)
-; SCALAR_EPILOGUE-NEXT:    [[TMP11:%.*]] = zext nneg <vscale x 16 x i32> [[TMP7]] to <vscale x 16 x i64>
-; SCALAR_EPILOGUE-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i8, ptr [[P]], <vscale x 16 x i64> [[TMP11]]
-; SCALAR_EPILOGUE-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 16 x i8> @llvm.masked.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> [[TMP12]], i32 1, <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i8> poison)
-; SCALAR_EPILOGUE-NEXT:    [[TMP13:%.*]] = zext nneg <vscale x 16 x i32> [[TMP8]] to <vscale x 16 x i64>
-; SCALAR_EPILOGUE-NEXT:    [[TMP14:%.*]] = getelementptr inbounds i8, ptr [[P]], <vscale x 16 x i64> [[TMP13]]
-; SCALAR_EPILOGUE-NEXT:    [[WIDE_MASKED_GATHER3:%.*]] = call <vscale x 16 x i8> @llvm.masked.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> [[TMP14]], i32 1, <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i8> poison)
-; SCALAR_EPILOGUE-NEXT:    [[TMP15:%.*]] = zext nneg <vscale x 16 x i32> [[TMP9]] to <vscale x 16 x i64>
-; SCALAR_EPILOGUE-NEXT:    [[TMP16:%.*]] = getelementptr inbounds i8, ptr [[P]], <vscale x 16 x i64> [[TMP15]]
-; SCALAR_EPILOGUE-NEXT:    [[WIDE_MASKED_GATHER4:%.*]] = call <vscale x 16 x i8> @llvm.masked.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> [[TMP16]], i32 1, <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i8> poison)
-; SCALAR_EPILOGUE-NEXT:    [[TMP17:%.*]] = zext nneg <vscale x 16 x i32> [[TMP10]] to <vscale x 16 x i64>
-; SCALAR_EPILOGUE-NEXT:    [[TMP18:%.*]] = getelementptr inbounds i8, ptr [[P]], <vscale x 16 x i64> [[TMP17]]
-; SCALAR_EPILOGUE-NEXT:    [[WIDE_MASKED_GATHER5:%.*]] = call <vscale x 16 x i8> @llvm.masked.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> [[TMP18]], i32 1, <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i8> poison)
-; SCALAR_EPILOGUE-NEXT:    [[TMP19:%.*]] = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> [[WIDE_MASKED_GATHER]], <vscale x 16 x i8> [[WIDE_MASKED_GATHER3]])
-; SCALAR_EPILOGUE-NEXT:    [[TMP20:%.*]] = sub <vscale x 16 x i8> zeroinitializer, [[TMP19]]
-; SCALAR_EPILOGUE-NEXT:    [[TMP21:%.*]] = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> [[WIDE_MASKED_GATHER4]], <vscale x 16 x i8> [[WIDE_MASKED_GATHER5]])
-; SCALAR_EPILOGUE-NEXT:    [[TMP22:%.*]] = sub <vscale x 16 x i8> zeroinitializer, [[TMP21]]
-; SCALAR_EPILOGUE-NEXT:    [[TMP23:%.*]] = zext nneg <vscale x 16 x i32> [[TMP7]] to <vscale x 16 x i64>
-; SCALAR_EPILOGUE-NEXT:    [[TMP24:%.*]] = getelementptr inbounds i8, ptr [[Q]], <vscale x 16 x i64> [[TMP23]]
-; SCALAR_EPILOGUE-NEXT:    call void @llvm.masked.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> [[TMP19]], <vscale x 16 x ptr> [[TMP24]], i32 1, <vscale x 16 x i1> [[TMP6]])
-; SCALAR_EPILOGUE-NEXT:    [[TMP25:%.*]] = zext nneg <vscale x 16 x i32> [[TMP8]] to <vscale x 16 x i64>
-; SCALAR_EPILOGUE-NEXT:    [[TMP26:%.*]] = getelementptr inbounds i8, ptr [[Q]], <vscale x 16 x i64> [[TMP25]]
-; SCALAR_EPILOGUE-NEXT:    call void @llvm.masked.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> [[TMP20]], <vscale x 16 x ptr> [[TMP26]], i32 1, <vscale x 16 x i1> [[TMP6]])
-; SCALAR_EPILOGUE-NEXT:    [[TMP27:%.*]] = zext nneg <vscale x 16 x i32> [[TMP9]] to <vscale x 16 x i64>
-; SCALAR_EPILOGUE-NEXT:    [[TMP28:%.*]] = getelementptr inbounds i8, ptr [[Q]], <vscale x 16 x i64> [[TMP27]]
-; SCALAR_EPILOGUE-NEXT:    call void @llvm.masked.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> [[TMP21]], <vscale x 16 x ptr> [[TMP28]], i32 1, <vscale x 16 x i1> [[TMP6]])
-; SCALAR_EPILOGUE-NEXT:    [[TMP29:%.*]] = zext nneg <vscale x 16 x i32> [[TMP10]] to <vscale x 16 x i64>
-; SCALAR_EPILOGUE-NEXT:    [[TMP30:%.*]] = getelementptr inbounds i8, ptr [[Q]], <vscale x 16 x i64> [[TMP29]]
-; SCALAR_EPILOGUE-NEXT:    call void @llvm.masked.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> [[TMP22]], <vscale x 16 x ptr> [[TMP30]], i32 1, <vscale x 16 x i1> [[TMP6]])
+; SCALAR_EPILOGUE-NEXT:    [[TMP7:%.*]] = shl i32 [[INDEX]], 2
+; SCALAR_EPILOGUE-NEXT:    [[TMP8:%.*]] = sext i32 [[TMP7]] to i64
+; SCALAR_EPILOGUE-NEXT:    [[TMP9:%.*]] = getelementptr i8, ptr [[P]], i64 [[TMP8]]
+; SCALAR_EPILOGUE-NEXT:    [[INTERLEAVED_MASK:%.*]] = call <vscale x 64 x i1> @llvm.vector.interleave4.nxv64i1(<vscale x 16 x i1> [[TMP6]], <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i1> [[TMP6]])
+; SCALAR_EPILOGUE-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 64 x i8> @llvm.masked.load.nxv64i8.p0(ptr [[TMP9]], i32 1, <vscale x 64 x i1> [[INTERLEAVED_MASK]], <vscale x 64 x i8> poison)
+; SCALAR_EPILOGUE-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.vector.deinterleave4.nxv64i8(<vscale x 64 x i8> [[WIDE_MASKED_VEC]])
+; SCALAR_EPILOGUE-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 0
+; SCALAR_EPILOGUE-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 1
+; SCALAR_EPILOGUE-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 2
+; SCALAR_EPILOGUE-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 3
+; SCALAR_EPILOGUE-NEXT:    [[TMP14:%.*]] = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> [[TMP10]], <vscale x 16 x i8> [[TMP11]])
+; SCALAR_EPILOGUE-NEXT:    [[TMP15:%.*]] = sub <vscale x 16 x i8> zeroinitializer, [[TMP14]]
+; SCALAR_EPILOGUE-NEXT:    [[TMP16:%.*]] = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> [[TMP12]], <vscale x 16 x i8> [[TMP13]])
+; SCALAR_EPILOGUE-NEXT:    [[TMP17:%.*]] = sub <vscale x 16 x i8> zeroinitializer, [[TMP16]]
+; SCALAR_EPILOGUE-NEXT:    [[TMP18:%.*]] = sext i32 [[TMP7]] to i64
+; SCALAR_EPILOGUE-NEXT:    [[TMP19:%.*]] = getelementptr i8, ptr [[Q]], i64 [[TMP18]]
+; SCALAR_EPILOGUE-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 64 x i8> @llvm.vector.interleave4.nxv64i8(<vscale x 16 x i8> [[TMP14]], <vscale x 16 x i8> [[TMP15]], <vscale x 16 x i8> [[TMP16]], <vscale x 16 x i8> [[TMP17]])
+; SCALAR_EPILOGUE-NEXT:    [[INTERLEAVED_MASK3:%.*]] = call <vscale x 64 x i1> @llvm.vector.interleave4.nxv64i1(<vscale x 16 x i1> [[TMP6]], <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i1> [[TMP6]])
+; SCALAR_EPILOGUE-NEXT:    call void @llvm.masked.store.nxv64i8.p0(<vscale x 64 x i8> [[INTERLEAVED_VEC]], ptr [[TMP19]], i32 1, <vscale x 64 x i1> [[INTERLEAVED_MASK3]])
 ; SCALAR_EPILOGUE-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[INDEX]], [[TMP4]]
 ; SCALAR_EPILOGUE-NEXT:    [[VEC_IND_NEXT]] = add <vscale x 16 x i32> [[VEC_IND]], [[BROADCAST_SPLAT2]]
-; SCALAR_EPILOGUE-NEXT:    [[TMP31:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
-; SCALAR_EPILOGUE-NEXT:    br i1 [[TMP31]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
+; SCALAR_EPILOGUE-NEXT:    [[TMP20:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
+; SCALAR_EPILOGUE-NEXT:    br i1 [[TMP20]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
 ; SCALAR_EPILOGUE:       middle.block:
 ; SCALAR_EPILOGUE-NEXT:    [[CMP_N:%.*]] = icmp eq i32 [[N_MOD_VF]], 0
 ; SCALAR_EPILOGUE-NEXT:    br i1 [[CMP_N]], label [[FOR_END:%.*]], label [[SCALAR_PH]]
@@ -281,42 +263,29 @@ define void @masked_strided_factor4(ptr noalias nocapture readonly %p, ptr noali
 ; PREDICATED_DATA-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <vscale x 16 x i1> @llvm.get.active.lane.mask.nxv16i1.i32(i32 [[INDEX]], i32 1024)
 ; PREDICATED_DATA-NEXT:    [[TMP5:%.*]] = icmp ugt <vscale x 16 x i32> [[VEC_IND]], [[BROADCAST_SPLAT]]
 ; PREDICATED_DATA-NEXT:    [[TMP6:%.*]] = select <vscale x 16 x i1> [[ACTIVE_LANE_MASK]], <vscale x 16 x i1> [[TMP5]], <vscale x 16 x i1> zeroinitializer
-; PREDICATED_DATA-NEXT:    [[TMP7:%.*]] = shl nuw nsw <vscale x 16 x i32> [[VEC_IND]], splat (i32 2)
-; PREDICATED_DATA-NEXT:    [[TMP8:%.*]] = or disjoint <vscale x 16 x i32> [[TMP7]], splat (i32 1)
-; PREDICATED_DATA-NEXT:    [[TMP9:%.*]] = or disjoint <vscale x 16 x i32> [[TMP7]], splat (i32 2)
-; PREDICATED_DATA-NEXT:    [[TMP10:%.*]] = or disjoint <vscale x 16 x i32> [[TMP7]], splat (i32 3)
-; PREDICATED_DATA-NEXT:    [[TMP11:%.*]] = zext nneg <vscale x 16 x i32> [[TMP7]] to <vscale x 16 x i64>
-; PREDICATED_DATA-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i8, ptr [[P]], <vscale x 16 x i64> [[TMP11]]
-; PREDICATED_DATA-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 16 x i8> @llvm.masked.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> [[TMP12]], i32 1, <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i8> poison)
-; PREDICATED_DATA-NEXT:    [[TMP13:%.*]] = zext nneg <vscale x 16 x i32> [[TMP8]] to <vscale x 16 x i64>
-; PREDICATED_DATA-NEXT:    [[TMP14:%.*]] = getelementptr inbounds i8, ptr [[P]], <vscale x 16 x i64> [[TMP13]]
-; PREDICATED_DATA-NEXT:    [[WIDE_MASKED_GATHER3:%.*]] = call <vscale x 16 x i8> @llvm.masked.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> [[TMP14]], i32 1, <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i8> poison)
-; PREDICATED_DATA-NEXT:    [[TMP15:%.*]] = zext nneg <vscale x 16 x i32> [[TMP9]] to <vscale x 16 x i64>
-; PREDICATED_DATA-NEXT:    [[TMP16:%.*]] = getelementptr inbounds i8, ptr [[P]], <vscale x 16 x i64> [[TMP15]]
-; PREDICATED_DATA-NEXT:    [[WIDE_MASKED_GATHER4:%.*]] = call <vscale x 16 x i8> @llvm.masked.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> [[TMP16]], i32 1, <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i8> poison)
-; PREDICATED_DATA-NEXT:    [[TMP17:%.*]] = zext nneg <vscale x 16 x i32> [[TMP10]] to <vscale x 16 x i64>
-; PREDICATED_DATA-NEXT:    [[TMP18:%.*]] = getelementptr inbounds i8, ptr [[P]], <vscale x 16 x i64> [[TMP17]]
-; PREDICATED_DATA-NEXT:    [[WIDE_MASKED_GATHER5:%.*]] = call <vscale x 16 x i8> @llvm.masked.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> [[TMP18]], i32 1, <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i8> poison)
-; PREDICATED_DATA-NEXT:    [[TMP19:%.*]] = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> [[WIDE_MASKED_GATHER]], <vscale x 16 x i8> [[WIDE_MASKED_GATHER3]])
-; PREDICATED_DATA-NEXT:    [[TMP20:%.*]] = sub <vscale x 16 x i8> zeroinitializer, [[TMP19]]
-; PREDICATED_DATA-NEXT:    [[TMP21:%.*]] = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> [[WIDE_MASKED_GATHER4]], <vscale x 16 x i8> [[WIDE_MASKED_GATHER5]])
-; PREDICATED_DATA-NEXT:    [[TMP22:%.*]] = sub <vscale x 16 x i8> zeroinitializer, [[TMP21]]
-; PREDICATED_DATA-NEXT:    [[TMP23:%.*]] = zext nneg <vscale x 16 x i32> [[TMP7]] to <vscale x 16 x i64>
-; PREDICATED_DATA-NEXT:    [[TMP24:%.*]] = getelementptr inbounds i8, ptr [[Q]], <vscale x 16 x i64> [[TMP23]]
-; PREDICATED_DATA-NEXT:    call void @llvm.masked.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> [[TMP19]], <vscale x 16 x ptr> [[TMP24]], i32 1, <vscale x 16 x i1> [[TMP6]])
-; PREDICATED_DATA-NEXT:    [[TMP25:%.*]] = zext nneg <vscale x 16 x i32> [[TMP8]] to <vscale x 16 x i64>
-; PREDICATED_DATA-NEXT:    [[TMP26:%.*]] = getelementptr inbounds i8, ptr [[Q]], <vscale x 16 x i64> [[TMP25]]
-; PREDICATED_DATA-NEXT:    call void @llvm.masked.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> [[TMP20]], <vscale x 16 x ptr> [[TMP26]], i32 1, <vscale x 16 x i1> [[TMP6]])
-; PREDICATED_DATA-NEXT:    [[TMP27:%.*]] = zext nneg <vscale x 16 x i32> [[TMP9]] to <vscale x 16 x i64>
-; PREDICATED_DATA-NEXT:    [[TMP28:%.*]] = getelementptr inbounds i8, ptr [[Q]], <vscale x 16 x i64> [[TMP27]]
-; PREDICATED_DATA-NEXT:    call void @llvm.masked.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> [[TMP21]], <vscale x 16 x ptr> [[TMP28]], i32 1, <vscale x 16 x i1> [[TMP6]])
-; PREDICATED_DATA-NEXT:    [[TMP29:%.*]] = zext nneg <vscale x 16 x i32> [[TMP10]] to <vscale x 16 x i64>
-; PREDICATED_DATA-NEXT:    [[TMP30:%.*]] = getelementptr inbounds i8, ptr [[Q]], <vscale x 16 x i64> [[TMP29]]
-; PREDICATED_DATA-NEXT:    call void @llvm.masked.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> [[TMP22]], <vscale x 16 x ptr> [[TMP30]], i32 1, <vscale x 16 x i1> [[TMP6]])
+; PREDICATED_DATA-NEXT:    [[TMP7:%.*]] = shl i32 [[INDEX]], 2
+; PREDICATED_DATA-NEXT:    [[TMP8:%.*]] = sext i32 [[TMP7]] to i64
+; PREDICATED_DATA-NEXT:    [[TMP9:%.*]] = getelementptr i8, ptr [[P]], i64 [[TMP8]]
+; PREDICATED_DATA-NEXT:    [[INTERLEAVED_MASK:%.*]] = call <vscale x 64 x i1> @llvm.vector.interleave4.nxv64i1(<vscale x 16 x i1> [[TMP6]], <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i1> [[TMP6]])
+; PREDICATED_DATA-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 64 x i8> @llvm.masked.load.nxv64i8.p0(ptr [[TMP9]], i32 1, <vscale x 64 x i1> [[INTERLEAVED_MASK]], <vscale x 64 x i8> poison)
+; PREDICATED_DATA-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.vector.deinterleave4.nxv64i8(<vscale x 64 x i8> [[WIDE_MASKED_VEC]])
+; PREDICATED_DATA-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 0
+; PREDICATED_DATA-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 1
+; PREDICATED_DATA-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 2
+; PREDICATED_DATA-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 3
+; PREDICATED_DATA-NEXT:    [[TMP14:%.*]] = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> [[TMP10]], <vscale x 16 x i8> [[TMP11]])
+; PREDICATED_DATA-NEXT:    [[TMP15:%.*]] = sub <vscale x 16 x i8> zeroinitializer, [[TMP14]]
+; PREDICATED_DATA-NEXT:    [[TMP16:%.*]] = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> [[TMP12]], <vscale x 16 x i8> [[TMP13]])
+; PREDICATED_DATA-NEXT:    [[TMP17:%.*]] = sub <vscale x 16 x i8> zeroinitializer, [[TMP16]]
+; PREDICATED_DATA-NEXT:    [[TMP18:%.*]] = sext i32 [[TMP7]] to i64
+; PREDICATED_DATA-NEXT:    [[TMP19:%.*]] = getelementptr i8, ptr [[Q]], i64 [[TMP18]]
+; PREDICATED_DATA-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 64 x i8> @llvm.vector.interleave4.nxv64i8(<vscale x 16 x i8> [[TMP14]], <vscale x 16 x i8> [[TMP15]], <vscale x 16 x i8> [[TMP16]], <vscale x 16 x i8> [[TMP17]])
+; PREDICATED_DATA-NEXT:    [[INTERLEAVED_MASK3:%.*]] = call <vscale x 64 x i1> @llvm.vector.interleave4.nxv64i1(<vscale x 16 x i1> [[TMP6]], <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i1> [[TMP6]])
+; PREDICATED_DATA-NEXT:    call void @llvm.masked.store.nxv64i8.p0(<vscale x 64 x i8> [[INTERLEAVED_VEC]], ptr [[TMP19]], i32 1, <vscale x 64 x i1> [[INTERLEAVED_MASK3]])
 ; PREDICATED_DATA-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[INDEX]], [[TMP3]]
 ; PREDICATED_DATA-NEXT:    [[VEC_IND_NEXT]] = add <vscale x 16 x i32> [[VEC_IND]], [[BROADCAST_SPLAT2]]
-; PREDICATED_DATA-NEXT:    [[TMP31:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
-; PREDICATED_DATA-NEXT:    br i1 [[TMP31]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
+; PREDICATED_DATA-NEXT:    [[TMP20:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
+; PREDICATED_DATA-NEXT:    br i1 [[TMP20]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
 ; PREDICATED_DATA:       middle.block:
 ; PREDICATED_DATA-NEXT:    br label [[FOR_END:%.*]]
 ; PREDICATED_DATA:       scalar.ph:
@@ -337,53 +306,38 @@ define void @masked_strided_factor4(ptr noalias nocapture readonly %p, ptr noali
 ; PREDICATED_DATA-WITH-EVL-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 16 x i32> poison, i32 [[CONV]], i64 0
 ; PREDICATED_DATA-WITH-EVL-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 16 x i32> [[BROADCAST_SPLATINSERT]], <vscale x 16 x i32> poison, <vscale x 16 x i32> zeroinitializer
 ; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP4:%.*]] = call <vscale x 16 x i32> @llvm.stepvector.nxv16i32()
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <vscale x 16 x i32> poison, i32 [[TMP3]], i64 0
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[BROADCAST_SPLAT2:%.*]] = shufflevector <vscale x 16 x i32> [[BROADCAST_SPLATINSERT1]], <vscale x 16 x i32> poison, <vscale x 16 x i32> zeroinitializer
 ; PREDICATED_DATA-WITH-EVL-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; PREDICATED_DATA-WITH-EVL:       vector.body:
 ; PREDICATED_DATA-WITH-EVL-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[EVL_BASED_IV:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_EVL_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; PREDICATED_DATA-WITH-EVL-NEXT:    [[VEC_IND:%.*]] = phi <vscale x 16 x i32> [ [[TMP4]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[AVL:%.*]] = sub i32 1024, [[EVL_BASED_IV]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP5:%.*]] = call i32 @llvm.experimental.get.vector.length.i32(i32 [[AVL]], i32 16, i1 true)
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <vscale x 16 x i32> poison, i32 [[TMP5]], i64 0
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[BROADCAST_SPLAT2:%.*]] = shufflevector <vscale x 16 x i32> [[BROADCAST_SPLATINSERT1]], <vscale x 16 x i32> poison, <vscale x 16 x i32> zeroinitializer
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP5:%.*]] = icmp ult <vscale x 16 x i32> [[VEC_IND]], splat (i32 1024)
 ; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP6:%.*]] = icmp ugt <vscale x 16 x i32> [[VEC_IND]], [[BROADCAST_SPLAT]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP7:%.*]] = shl nuw nsw <vscale x 16 x i32> [[VEC_IND]], splat (i32 2)
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP8:%.*]] = or disjoint <vscale x 16 x i32> [[TMP7]], splat (i32 1)
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP9:%.*]] = or disjoint <vscale x 16 x i32> [[TMP7]], splat (i32 2)
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP10:%.*]] = or disjoint <vscale x 16 x i32> [[TMP7]], splat (i32 3)
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP11:%.*]] = zext nneg <vscale x 16 x i32> [[TMP7]] to <vscale x 16 x i64>
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i8, ptr [[P]], <vscale x 16 x i64> [[TMP11]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 16 x i8> @llvm.vp.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> align 1 [[TMP12]], <vscale x 16 x i1> [[TMP6]], i32 [[TMP5]])
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP13:%.*]] = zext nneg <vscale x 16 x i32> [[TMP8]] to <vscale x 16 x i64>
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP14:%.*]] = getelementptr inbounds i8, ptr [[P]], <vscale x 16 x i64> [[TMP13]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[WIDE_MASKED_GATHER3:%.*]] = call <vscale x 16 x i8> @llvm.vp.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> align 1 [[TMP14]], <vscale x 16 x i1> [[TMP6]], i32 [[TMP5]])
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP15:%.*]] = zext nneg <vscale x 16 x i32> [[TMP9]] to <vscale x 16 x i64>
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP16:%.*]] = getelementptr inbounds i8, ptr [[P]], <vscale x 16 x i64> [[TMP15]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[WIDE_MASKED_GATHER4:%.*]] = call <vscale x 16 x i8> @llvm.vp.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> align 1 [[TMP16]], <vscale x 16 x i1> [[TMP6]], i32 [[TMP5]])
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP17:%.*]] = zext nneg <vscale x 16 x i32> [[TMP10]] to <vscale x 16 x i64>
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP18:%.*]] = getelementptr inbounds i8, ptr [[P]], <vscale x 16 x i64> [[TMP17]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[WIDE_MASKED_GATHER5:%.*]] = call <vscale x 16 x i8> @llvm.vp.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> align 1 [[TMP18]], <vscale x 16 x i1> [[TMP6]], i32 [[TMP5]])
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP19:%.*]] = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> [[WIDE_MASKED_GATHER]], <vscale x 16 x i8> [[WIDE_MASKED_GATHER3]])
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP20:%.*]] = sub <vscale x 16 x i8> zeroinitializer, [[TMP19]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP21:%.*]] = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> [[WIDE_MASKED_GATHER4]], <vscale x 16 x i8> [[WIDE_MASKED_GATHER5]])
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP22:%.*]] = sub <vscale x 16 x i8> zeroinitializer, [[TMP21]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP23:%.*]] = zext nneg <vscale x 16 x i32> [[TMP7]] to <vscale x 16 x i64>
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP24:%.*]] = getelementptr inbounds i8, ptr [[Q]], <vscale x 16 x i64> [[TMP23]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    call void @llvm.vp.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> [[TMP19]], <vscale x 16 x ptr> align 1 [[TMP24]], <vscale x 16 x i1> [[TMP6]], i32 [[TMP5]])
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP25:%.*]] = zext nneg <vscale x 16 x i32> [[TMP8]] to <vscale x 16 x i64>
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP26:%.*]] = getelementptr inbounds i8, ptr [[Q]], <vscale x 16 x i64> [[TMP25]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    call void @llvm.vp.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> [[TMP20]], <vscale x 16 x ptr> align 1 [[TMP26]], <vscale x 16 x i1> [[TMP6]], i32 [[TMP5]])
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP27:%.*]] = zext nneg <vscale x 16 x i32> [[TMP9]] to <vscale x 16 x i64>
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP28:%.*]] = getelementptr inbounds i8, ptr [[Q]], <vscale x 16 x i64> [[TMP27]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    call void @llvm.vp.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> [[TMP21]], <vscale x 16 x ptr> align 1 [[TMP28]], <vscale x 16 x i1> [[TMP6]], i32 [[TMP5]])
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP29:%.*]] = zext nneg <vscale x 16 x i32> [[TMP10]] to <vscale x 16 x i64>
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP30:%.*]] = getelementptr inbounds i8, ptr [[Q]], <vscale x 16 x i64> [[TMP29]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    call void @llvm.vp.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> [[TMP22]], <vscale x 16 x ptr> align 1 [[TMP30]], <vscale x 16 x i1> [[TMP6]], i32 [[TMP5]])
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i32 [[TMP5]], [[EVL_BASED_IV]]
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP7:%.*]] = select <vscale x 16 x i1> [[TMP5]], <vscale x 16 x i1> [[TMP6]], <vscale x 16 x i1> zeroinitializer
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP8:%.*]] = shl i32 [[INDEX]], 2
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP9:%.*]] = sext i32 [[TMP8]] to i64
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP10:%.*]] = getelementptr i8, ptr [[P]], i64 [[TMP9]]
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[INTERLEAVED_MASK:%.*]] = call <vscale x 64 x i1> @llvm.vector.interleave4.nxv64i1(<vscale x 16 x i1> [[TMP7]], <vscale x 16 x i1> [[TMP7]], <vscale x 16 x i1> [[TMP7]], <vscale x 16 x i1> [[TMP7]])
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 64 x i8> @llvm.masked.load.nxv64i8.p0(ptr [[TMP10]], i32 1, <vscale x 64 x i1> [[INTERLEAVED_MASK]], <vscale x 64 x i8> poison)
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.vector.deinterleave4.nxv64i8(<vscale x 64 x i8> [[WIDE_MASKED_VEC]])
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 0
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 1
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 2
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 3
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP15:%.*]] = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> [[TMP11]], <vscale x 16 x i8> [[TMP12]])
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP16:%.*]] = sub <vscale x 16 x i8> zeroinitializer, [[TMP15]]
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP17:%.*]] = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> [[TMP13]], <vscale x 16 x i8> [[TMP14]])
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP18:%.*]] = sub <vscale x 16 x i8> zeroinitializer, [[TMP17]]
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP19:%.*]] = sext i32 [[TMP8]] to i64
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP20:%.*]] = getelementptr i8, ptr [[Q]], i64 [[TMP19]]
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 64 x i8> @llvm.vector.interleave4.nxv64i8(<vscale x 16 x i8> [[TMP15]], <vscale x 16 x i8> [[TMP16]], <vscale x 16 x i8> [[TMP17]], <vscale x 16 x i8> [[TMP18]])
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[INTERLEAVED_MASK3:%.*]] = call <vscale x 64 x i1> @llvm.vector.interleave4.nxv64i1(<vscale x 16 x i1> [[TMP7]], <vscale x 16 x i1> [[TMP7]], <vscale x 16 x i1> [[TMP7]], <vscale x 16 x i1> [[TMP7]])
+; PREDICATED_DATA-WITH-EVL-NEXT:    call void @llvm.masked.store.nxv64i8.p0(<vscale x 64 x i8> [[INTERLEAVED_VEC]], ptr [[TMP20]], i32 1, <vscale x 64 x i1> [[INTERLEAVED_MASK3]])
 ; PREDICATED_DATA-WITH-EVL-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[INDEX]], [[TMP3]]
 ; PREDICATED_DATA-WITH-EVL-NEXT:    [[VEC_IND_NEXT]] = add <vscale x 16 x i32> [[VEC_IND]], [[BROADCAST_SPLAT2]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP31:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
-; PREDICATED_DATA-WITH-EVL-NEXT:    br i1 [[TMP31]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
+; PREDICATED_DATA-WITH-EVL-NEXT:    [[TMP21:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
+; PREDICATED_DATA-WITH-EVL-NEXT:    br i1 [[TMP21]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
 ; PREDICATED_DATA-WITH-EVL:       middle.block:
 ; PREDICATED_DATA-WITH-EVL-NEXT:    br label [[FOR_END:%.*]]
 ; PREDICATED_DATA-WITH-EVL:       scalar.ph:
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-interleave.ll b/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-interleave.ll
index acfcf90..556446c3 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-interleave.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-interleave.ll
@@ -21,36 +21,33 @@ define void @interleave(ptr noalias %a, ptr noalias %b, i64 %N) {
 ; IF-EVL-NEXT:    [[N_RND_UP:%.*]] = add i64 [[N:%.*]], [[TMP6]]
 ; IF-EVL-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP5]]
 ; IF-EVL-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
+; IF-EVL-NEXT:    [[TRIP_COUNT_MINUS_1:%.*]] = sub i64 [[N]], 1
 ; IF-EVL-NEXT:    [[TMP7:%.*]] = call i64 @llvm.vscale.i64()
 ; IF-EVL-NEXT:    [[TMP8:%.*]] = mul nuw i64 [[TMP7]], 4
-; IF-EVL-NEXT:    [[TMP10:%.*]] = call <vscale x 4 x i64> @llvm.stepvector.nxv4i64()
-; IF-EVL-NEXT:    [[TMP12:%.*]] = mul <vscale x 4 x i64> [[TMP10]], splat (i64 1)
-; IF-EVL-NEXT:    [[INDUCTION:%.*]] = add <vscale x 4 x i64> zeroinitializer, [[TMP12]]
+; IF-EVL-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[TRIP_COUNT_MINUS_1]], i64 0
+; IF-EVL-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 4 x i64> [[BROADCAST_SPLATINSERT]], <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
 ; IF-EVL-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; IF-EVL:       vector.body:
-; IF-EVL-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; IF-EVL-NEXT:    [[EVL_BASED_IV:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_EVL_NEXT:%.*]], [[VECTOR_BODY]] ]
-; IF-EVL-NEXT:    [[VEC_IND:%.*]] = phi <vscale x 4 x i64> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
-; IF-EVL-NEXT:    [[AVL:%.*]] = sub i64 [[N]], [[EVL_BASED_IV]]
-; IF-EVL-NEXT:    [[TMP11:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 4, i1 true)
-; IF-EVL-NEXT:    [[TMP13:%.*]] = zext i32 [[TMP11]] to i64
-; IF-EVL-NEXT:    [[TMP9:%.*]] = mul i64 1, [[TMP13]]
-; IF-EVL-NEXT:    [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[TMP9]], i64 0
+; IF-EVL-NEXT:    [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[EVL_BASED_IV]], i64 0
 ; IF-EVL-NEXT:    [[BROADCAST_SPLAT2:%.*]] = shufflevector <vscale x 4 x i64> [[BROADCAST_SPLATINSERT1]], <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
-; IF-EVL-NEXT:    [[TMP21:%.*]] = getelementptr inbounds [2 x i32], ptr [[B:%.*]], <vscale x 4 x i64> [[VEC_IND]], i32 0
-; IF-EVL-NEXT:    [[WIDE_MASKED_GATHER3:%.*]] = call <vscale x 4 x i32> @llvm.vp.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> align 4 [[TMP21]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP11]])
-; IF-EVL-NEXT:    [[TMP23:%.*]] = getelementptr inbounds [2 x i32], ptr [[B]], <vscale x 4 x i64> [[VEC_IND]], i32 1
-; IF-EVL-NEXT:    [[WIDE_MASKED_GATHER5:%.*]] = call <vscale x 4 x i32> @llvm.vp.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> align 4 [[TMP23]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP11]])
+; IF-EVL-NEXT:    [[TMP9:%.*]] = call <vscale x 4 x i64> @llvm.stepvector.nxv4i64()
+; IF-EVL-NEXT:    [[TMP10:%.*]] = add <vscale x 4 x i64> zeroinitializer, [[TMP9]]
+; IF-EVL-NEXT:    [[VEC_IV:%.*]] = add <vscale x 4 x i64> [[BROADCAST_SPLAT2]], [[TMP10]]
+; IF-EVL-NEXT:    [[TMP11:%.*]] = icmp ule <vscale x 4 x i64> [[VEC_IV]], [[BROADCAST_SPLAT]]
+; IF-EVL-NEXT:    [[TMP12:%.*]] = getelementptr inbounds [2 x i32], ptr [[B:%.*]], i64 [[EVL_BASED_IV]], i32 0
+; IF-EVL-NEXT:    [[INTERLEAVED_MASK:%.*]] = call <vscale x 8 x i1> @llvm.vector.interleave2.nxv8i1(<vscale x 4 x i1> [[TMP11]], <vscale x 4 x i1> [[TMP11]])
+; IF-EVL-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 8 x i32> @llvm.masked.load.nxv8i32.p0(ptr [[TMP12]], i32 4, <vscale x 8 x i1> [[INTERLEAVED_MASK]], <vscale x 8 x i32> poison)
+; IF-EVL-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_MASKED_VEC]])
+; IF-EVL-NEXT:    [[WIDE_MASKED_GATHER3:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
+; IF-EVL-NEXT:    [[WIDE_MASKED_GATHER5:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 1
 ; IF-EVL-NEXT:    [[TMP26:%.*]] = add nsw <vscale x 4 x i32> [[WIDE_MASKED_GATHER5]], [[WIDE_MASKED_GATHER3]]
 ; IF-EVL-NEXT:    [[TMP27:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[TMP29:%.*]] = getelementptr inbounds i32, ptr [[TMP27]], i32 0
-; IF-EVL-NEXT:    call void @llvm.vp.store.nxv4i32.p0(<vscale x 4 x i32> [[TMP26]], ptr align 4 [[TMP29]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP11]])
-; IF-EVL-NEXT:    [[TMP15:%.*]] = zext i32 [[TMP11]] to i64
-; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add i64 [[TMP15]], [[EVL_BASED_IV]]
-; IF-EVL-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP8]]
-; IF-EVL-NEXT:    [[VEC_IND_NEXT]] = add <vscale x 4 x i64> [[VEC_IND]], [[BROADCAST_SPLAT2]]
-; IF-EVL-NEXT:    [[TMP33:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; IF-EVL-NEXT:    br i1 [[TMP33]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; IF-EVL-NEXT:    call void @llvm.masked.store.nxv4i32.p0(<vscale x 4 x i32> [[TMP26]], ptr [[TMP29]], i32 4, <vscale x 4 x i1> [[TMP11]])
+; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add i64 [[EVL_BASED_IV]], [[TMP8]]
+; IF-EVL-NEXT:    [[TMP14:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], [[N_VEC]]
+; IF-EVL-NEXT:    br i1 [[TMP14]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
 ; IF-EVL:       middle.block:
 ; IF-EVL-NEXT:    br label [[FOR_COND_CLEANUP:%.*]]
 ; IF-EVL:       scalar.ph:
@@ -67,7 +64,7 @@ define void @interleave(ptr noalias %a, ptr noalias %b, i64 %N) {
 ; IF-EVL-NEXT:    store i32 [[ADD]], ptr [[ARRAYIDX4]], align 4
 ; IF-EVL-NEXT:    [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
 ; IF-EVL-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i64 [[IV_NEXT]], [[N]]
-; IF-EVL-NEXT:    br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
 ; IF-EVL:       for.cond.cleanup:
 ; IF-EVL-NEXT:    ret void
 ;