diff --git a/llvm/lib/Target/LoongArch/LoongArchISelLowering.cpp b/llvm/lib/Target/LoongArch/LoongArchISelLowering.cpp
index d4e1d9c6f3ca6..1383302059910 100644
--- a/llvm/lib/Target/LoongArch/LoongArchISelLowering.cpp
+++ b/llvm/lib/Target/LoongArch/LoongArchISelLowering.cpp
@@ -31,6 +31,7 @@
 #include "llvm/Support/ErrorHandling.h"
 #include "llvm/Support/KnownBits.h"
 #include "llvm/Support/MathExtras.h"
+#include <llvm/Analysis/VectorUtils.h>
 
 using namespace llvm;
 
@@ -543,6 +544,37 @@ SDValue LoongArchTargetLowering::lowerBITREVERSE(SDValue Op,
   }
 }
 
+// Widen element type to get a new mask value (if possible).
+// For example:
+//  shufflevector <4 x i32> %a, <4 x i32> %b,
+//                <4 x i32> <i32 6, i32 7, i32 2, i32 3>
+// is equivalent to:
+//  shufflevector <2 x i64> %a, <2 x i64> %b, <2 x i32> <i32 3, i32 1>
+// can be lowered to:
+//  VPACKOD_D vr0, vr0, vr1
+static SDValue widenShuffleMask(const SDLoc &DL, ArrayRef<int> Mask, MVT VT,
+                                SDValue V1, SDValue V2, SelectionDAG &DAG) {
+  unsigned EltBits = VT.getScalarSizeInBits();
+
+  if (EltBits > 32 || EltBits == 1)
+    return SDValue();
+
+  SmallVector<int, 8> NewMask;
+  if (widenShuffleMaskElts(Mask, NewMask)) {
+    MVT NewEltVT = VT.isFloatingPoint() ? MVT::getFloatingPointVT(EltBits * 2)
+                                        : MVT::getIntegerVT(EltBits * 2);
+    MVT NewVT = MVT::getVectorVT(NewEltVT, VT.getVectorNumElements() / 2);
+    if (DAG.getTargetLoweringInfo().isTypeLegal(NewVT)) {
+      SDValue NewV1 = DAG.getBitcast(NewVT, V1);
+      SDValue NewV2 = DAG.getBitcast(NewVT, V2);
+      return DAG.getBitcast(
+          VT, DAG.getVectorShuffle(NewVT, DL, NewV1, NewV2, NewMask));
+    }
+  }
+
+  return SDValue();
+}
+
 /// Attempts to match a shuffle mask against the VBSLL, VBSRL, VSLLI and VSRLI
 /// instruction.
 // The funciton matches elements from one of the input vector shuffled to the
@@ -1365,6 +1397,8 @@ static SDValue lower128BitShuffle(const SDLoc &DL, ArrayRef<int> Mask, MVT VT,
     return Result;
   if ((Result = lowerVECTOR_SHUFFLEAsByteRotate(DL, Mask, VT, V1, V2, DAG)))
     return Result;
+  if (SDValue NewShuffle = widenShuffleMask(DL, Mask, VT, V1, V2, DAG))
+    return NewShuffle;
   if ((Result = lowerVECTOR_SHUFFLE_VSHUF(DL, Mask, VT, V1, V2, DAG)))
     return Result;
   return SDValue();
@@ -1803,6 +1837,8 @@ static SDValue lower256BitShuffle(const SDLoc &DL, ArrayRef<int> Mask, MVT VT,
     return Result;
   if ((Result = lowerVECTOR_SHUFFLEAsByteRotate(DL, NewMask, VT, V1, V2, DAG)))
     return Result;
+  if (SDValue NewShuffle = widenShuffleMask(DL, NewMask, VT, V1, V2, DAG))
+    return NewShuffle;
   if ((Result = lowerVECTOR_SHUFFLE_XVSHUF(DL, NewMask, VT, V1, V2, DAG)))
     return Result;
 
diff --git a/llvm/test/CodeGen/LoongArch/lasx/widen-shuffle-mask.ll b/llvm/test/CodeGen/LoongArch/lasx/widen-shuffle-mask.ll
index c32a60622f2a1..df639cb78cd1f 100644
--- a/llvm/test/CodeGen/LoongArch/lasx/widen-shuffle-mask.ll
+++ b/llvm/test/CodeGen/LoongArch/lasx/widen-shuffle-mask.ll
@@ -6,7 +6,8 @@ define <32 x i8> @widen_shuffle_mask_v32i8_to_v16i16(<32 x i8> %a, <32 x i8> %b)
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI0_0)
 ; CHECK-NEXT:    xvld $xr2, $a0, %pc_lo12(.LCPI0_0)
-; CHECK-NEXT:    xvshuf.b $xr0, $xr1, $xr0, $xr2
+; CHECK-NEXT:    xvshuf.h $xr2, $xr1, $xr0
+; CHECK-NEXT:    xvori.b $xr0, $xr2, 0
 ; CHECK-NEXT:    ret
     %r = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 32, i32 33, i32 34, i32 35, i32 24, i32 25, i32 26, i32 27, i32 48, i32 49, i32 50, i32 51, i32 52, i32 53, i32 54, i32 55, i32 60, i32 61, i32 30, i32 31>
     ret <32 x i8> %r
@@ -17,7 +18,8 @@ define <32 x i8> @widen_shuffle_mask_v32i8_to_v8i32(<32 x i8> %a, <32 x i8> %b)
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI1_0)
 ; CHECK-NEXT:    xvld $xr2, $a0, %pc_lo12(.LCPI1_0)
-; CHECK-NEXT:    xvshuf.b $xr0, $xr1, $xr0, $xr2
+; CHECK-NEXT:    xvshuf.w $xr2, $xr1, $xr0
+; CHECK-NEXT:    xvori.b $xr0, $xr2, 0
 ; CHECK-NEXT:    ret
     %r = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 32, i32 33, i32 34, i32 35, i32 24, i32 25, i32 26, i32 27, i32 48, i32 49, i32 50, i32 51, i32 52, i32 53, i32 54, i32 55, i32 60, i32 61, i32 62, i32 63>
     ret <32 x i8> %r
@@ -28,7 +30,8 @@ define <32 x i8> @widen_shuffle_mask_v32i8_to_v4i64(<32 x i8> %a, <32 x i8> %b)
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI2_0)
 ; CHECK-NEXT:    xvld $xr2, $a0, %pc_lo12(.LCPI2_0)
-; CHECK-NEXT:    xvshuf.b $xr0, $xr1, $xr0, $xr2
+; CHECK-NEXT:    xvshuf.d $xr2, $xr1, $xr0
+; CHECK-NEXT:    xvori.b $xr0, $xr2, 0
 ; CHECK-NEXT:    ret
     %r = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31, i32 48, i32 49, i32 50, i32 51, i32 52, i32 53, i32 54, i32 55>
     ret <32 x i8> %r
@@ -39,7 +42,7 @@ define <16 x i16> @widen_shuffle_mask_v16i16_to_v8i32(<16 x i16> %a, <16 x i16>
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI3_0)
 ; CHECK-NEXT:    xvld $xr2, $a0, %pc_lo12(.LCPI3_0)
-; CHECK-NEXT:    xvshuf.h $xr2, $xr1, $xr0
+; CHECK-NEXT:    xvshuf.w $xr2, $xr1, $xr0
 ; CHECK-NEXT:    xvori.b $xr0, $xr2, 0
 ; CHECK-NEXT:    ret
     %r = shufflevector <16 x i16> %a, <16 x i16> %b, <16 x i32> <i32 0, i32 1, i32 6, i32 7, i32 16, i32 17, i32 2, i32 3, i32 10, i32 11, i32 12, i32 13, i32 24, i32 25, i32 26, i32 27>
@@ -51,7 +54,7 @@ define <16 x i16> @widen_shuffle_mask_v16i16_to_v4i64(<16 x i16> %a, <16 x i16>
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI4_0)
 ; CHECK-NEXT:    xvld $xr2, $a0, %pc_lo12(.LCPI4_0)
-; CHECK-NEXT:    xvshuf.h $xr2, $xr1, $xr0
+; CHECK-NEXT:    xvshuf.d $xr2, $xr1, $xr0
 ; CHECK-NEXT:    xvori.b $xr0, $xr2, 0
 ; CHECK-NEXT:    ret
     %r = shufflevector <16 x i16> %a, <16 x i16> %b, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 20, i32 21, i32 22, i32 23, i32 12, i32 13, i32 14, i32 15, i32 28, i32 29, i32 30, i32 31>
@@ -63,7 +66,7 @@ define <8 x i32> @widen_shuffle_mask_v8i32_to_v4i64(<8 x i32> %a, <8 x i32> %b)
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI5_0)
 ; CHECK-NEXT:    xvld $xr2, $a0, %pc_lo12(.LCPI5_0)
-; CHECK-NEXT:    xvshuf.w $xr2, $xr1, $xr0
+; CHECK-NEXT:    xvshuf.d $xr2, $xr1, $xr0
 ; CHECK-NEXT:    xvori.b $xr0, $xr2, 0
 ; CHECK-NEXT:    ret
     %r = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 1, i32 8, i32 9, i32 14, i32 15, i32 6, i32 7>
@@ -73,9 +76,7 @@ define <8 x i32> @widen_shuffle_mask_v8i32_to_v4i64(<8 x i32> %a, <8 x i32> %b)
 define <32 x i8> @widen_shuffle_mask_v32i8_to_xvpackev_h(<32 x i8> %a, <32 x i8> %b) {
 ; CHECK-LABEL: widen_shuffle_mask_v32i8_to_xvpackev_h:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI6_0)
-; CHECK-NEXT:    xvld $xr2, $a0, %pc_lo12(.LCPI6_0)
-; CHECK-NEXT:    xvshuf.b $xr0, $xr1, $xr0, $xr2
+; CHECK-NEXT:    xvpackev.h $xr0, $xr1, $xr0
 ; CHECK-NEXT:    ret
     %r = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 0, i32 1, i32 32, i32 33, i32 4, i32 5, i32 36, i32 37, i32 8, i32 9, i32 40, i32 41, i32 12, i32 13, i32 44, i32 45, i32 16, i32 17, i32 48, i32 49, i32 20, i32 21, i32 52, i32 53, i32 24, i32 25, i32 56, i32 57, i32 28, i32 29, i32 60, i32 61>
     ret <32 x i8> %r
@@ -84,9 +85,7 @@ define <32 x i8> @widen_shuffle_mask_v32i8_to_xvpackev_h(<32 x i8> %a, <32 x i8>
 define <32 x i8> @widen_shuffle_mask_v32i8_to_xvpackod_h(<32 x i8> %a, <32 x i8> %b) {
 ; CHECK-LABEL: widen_shuffle_mask_v32i8_to_xvpackod_h:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI7_0)
-; CHECK-NEXT:    xvld $xr2, $a0, %pc_lo12(.LCPI7_0)
-; CHECK-NEXT:    xvshuf.b $xr0, $xr1, $xr0, $xr2
+; CHECK-NEXT:    xvpackod.h $xr0, $xr1, $xr0
 ; CHECK-NEXT:    ret
     %r = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 2, i32 3, i32 34, i32 35, i32 6, i32 7, i32 38, i32 39, i32 10, i32 11, i32 42, i32 43, i32 14, i32 15, i32 46, i32 47, i32 18, i32 19, i32 50, i32 51, i32 22, i32 23, i32 54, i32 55, i32 26, i32 27, i32 58, i32 59, i32 30, i32 31, i32 62, i32 63>
     ret <32 x i8> %r
@@ -95,9 +94,7 @@ define <32 x i8> @widen_shuffle_mask_v32i8_to_xvpackod_h(<32 x i8> %a, <32 x i8>
 define <32 x i8> @widen_shuffle_mask_v32i8_to_xvpickev_h(<32 x i8> %a, <32 x i8> %b) {
 ; CHECK-LABEL: widen_shuffle_mask_v32i8_to_xvpickev_h:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI8_0)
-; CHECK-NEXT:    xvld $xr2, $a0, %pc_lo12(.LCPI8_0)
-; CHECK-NEXT:    xvshuf.b $xr0, $xr1, $xr0, $xr2
+; CHECK-NEXT:    xvpickev.h $xr0, $xr1, $xr0
 ; CHECK-NEXT:    ret
     %r = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 0, i32 1, i32 4, i32 5, i32 8, i32 9, i32 12, i32 13, i32 32, i32 33, i32 36, i32 37, i32 40, i32 41, i32 44, i32 45, i32 16, i32 17, i32 20, i32 21, i32 24, i32 25, i32 28, i32 29, i32 48, i32 49, i32 52, i32 53, i32 56, i32 57, i32 60, i32 61>
     ret <32 x i8> %r
@@ -106,9 +103,7 @@ define <32 x i8> @widen_shuffle_mask_v32i8_to_xvpickev_h(<32 x i8> %a, <32 x i8>
 define <32 x i8> @widen_shuffle_mask_v32i8_to_xvpickod_h(<32 x i8> %a, <32 x i8> %b) {
 ; CHECK-LABEL: widen_shuffle_mask_v32i8_to_xvpickod_h:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI9_0)
-; CHECK-NEXT:    xvld $xr2, $a0, %pc_lo12(.LCPI9_0)
-; CHECK-NEXT:    xvshuf.b $xr0, $xr1, $xr0, $xr2
+; CHECK-NEXT:    xvpickod.h $xr0, $xr1, $xr0
 ; CHECK-NEXT:    ret
     %r = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 2, i32 3, i32 6, i32 7, i32 10, i32 11, i32 14, i32 15, i32 34, i32 35, i32 38, i32 39, i32 42, i32 43, i32 46, i32 47, i32 18, i32 19, i32 22, i32 23, i32 26, i32 27, i32 30, i32 31, i32 50, i32 51, i32 54, i32 55, i32 58, i32 59, i32 62, i32 63>
     ret <32 x i8> %r
@@ -117,9 +112,7 @@ define <32 x i8> @widen_shuffle_mask_v32i8_to_xvpickod_h(<32 x i8> %a, <32 x i8>
 define <32 x i8> @widen_shuffle_mask_v32i8_to_xvilvl_h(<32 x i8> %a, <32 x i8> %b) {
 ; CHECK-LABEL: widen_shuffle_mask_v32i8_to_xvilvl_h:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI10_0)
-; CHECK-NEXT:    xvld $xr2, $a0, %pc_lo12(.LCPI10_0)
-; CHECK-NEXT:    xvshuf.b $xr0, $xr1, $xr0, $xr2
+; CHECK-NEXT:    xvilvl.h $xr0, $xr1, $xr0
 ; CHECK-NEXT:    ret
     %r = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 0, i32 1, i32 32, i32 33, i32 2, i32 3, i32 34, i32 35, i32 4, i32 5, i32 36, i32 37, i32 6, i32 7, i32 38, i32 39, i32 16, i32 17, i32 48, i32 49, i32 18, i32 19, i32 50, i32 51, i32 20, i32 21, i32 52, i32 53, i32 22, i32 23, i32 54, i32 55>
     ret <32 x i8> %r
@@ -128,9 +121,7 @@ define <32 x i8> @widen_shuffle_mask_v32i8_to_xvilvl_h(<32 x i8> %a, <32 x i8> %
 define <32 x i8> @widen_shuffle_mask_v32i8_to_xvilvh_h(<32 x i8> %a, <32 x i8> %b) {
 ; CHECK-LABEL: widen_shuffle_mask_v32i8_to_xvilvh_h:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI11_0)
-; CHECK-NEXT:    xvld $xr2, $a0, %pc_lo12(.LCPI11_0)
-; CHECK-NEXT:    xvshuf.b $xr0, $xr1, $xr0, $xr2
+; CHECK-NEXT:    xvilvh.h $xr0, $xr1, $xr0
 ; CHECK-NEXT:    ret
     %r = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 8, i32 9, i32 40, i32 41, i32 10, i32 11, i32 42, i32 43, i32 12, i32 13, i32 44, i32 45, i32 14, i32 15, i32 46, i32 47, i32 24, i32 25, i32 56, i32 57, i32 26, i32 27, i32 58, i32 59, i32 28, i32 29, i32 60, i32 61, i32 30, i32 31, i32 62, i32 63>
     ret <32 x i8> %r
diff --git a/llvm/test/CodeGen/LoongArch/lsx/widen-shuffle-mask.ll b/llvm/test/CodeGen/LoongArch/lsx/widen-shuffle-mask.ll
index 35457ffa59586..bd3b7d0951565 100644
--- a/llvm/test/CodeGen/LoongArch/lsx/widen-shuffle-mask.ll
+++ b/llvm/test/CodeGen/LoongArch/lsx/widen-shuffle-mask.ll
@@ -6,7 +6,8 @@ define <16 x i8> @widen_shuffle_mask_v16i8_to_v8i16(<16 x i8> %a, <16 x i8> %b)
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI0_0)
 ; CHECK-NEXT:    vld $vr2, $a0, %pc_lo12(.LCPI0_0)
-; CHECK-NEXT:    vshuf.b $vr0, $vr1, $vr0, $vr2
+; CHECK-NEXT:    vshuf.h $vr2, $vr1, $vr0
+; CHECK-NEXT:    vori.b $vr0, $vr2, 0
 ; CHECK-NEXT:    ret
     %r = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 0, i32 1, i32 24, i32 25, i32 26, i32 27, i32 8, i32 9, i32 10, i32 11, i32 16, i32 17, i32 18, i32 19, i32 2, i32 3>
     ret <16 x i8> %r
@@ -17,7 +18,8 @@ define <16 x i8> @widen_shuffle_mask_v16i8_to_v4i32(<16 x i8> %a, <16 x i8> %b)
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI1_0)
 ; CHECK-NEXT:    vld $vr2, $a0, %pc_lo12(.LCPI1_0)
-; CHECK-NEXT:    vshuf.b $vr0, $vr1, $vr0, $vr2
+; CHECK-NEXT:    vshuf.w $vr2, $vr1, $vr0
+; CHECK-NEXT:    vori.b $vr0, $vr2, 0
 ; CHECK-NEXT:    ret
     %r = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 24, i32 25, i32 26, i32 27, i32 8, i32 9, i32 10, i32 11, i32 16, i32 17, i32 18, i32 19>
     ret <16 x i8> %r
@@ -28,7 +30,8 @@ define <16 x i8> @widen_shuffle_mask_v16i8_to_v2i64(<16 x i8> %a, <16 x i8> %b)
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI2_0)
 ; CHECK-NEXT:    vld $vr2, $a0, %pc_lo12(.LCPI2_0)
-; CHECK-NEXT:    vshuf.b $vr0, $vr1, $vr0, $vr2
+; CHECK-NEXT:    vshuf.d $vr2, $vr1, $vr0
+; CHECK-NEXT:    vori.b $vr0, $vr2, 0
 ; CHECK-NEXT:    ret
     %r = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
     ret <16 x i8> %r
@@ -39,7 +42,7 @@ define <8 x i16> @widen_shuffle_mask_v8i16_to_v4i32(<8 x i16> %a, <8 x i16> %b)
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI3_0)
 ; CHECK-NEXT:    vld $vr2, $a0, %pc_lo12(.LCPI3_0)
-; CHECK-NEXT:    vshuf.h $vr2, $vr1, $vr0
+; CHECK-NEXT:    vshuf.w $vr2, $vr1, $vr0
 ; CHECK-NEXT:    vori.b $vr0, $vr2, 0
 ; CHECK-NEXT:    ret
     %r = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 1, i32 12, i32 13, i32 14, i32 15, i32 2, i32 3>
@@ -51,7 +54,7 @@ define <8 x i16> @widen_shuffle_mask_v8i16_to_v2i64(<8 x i16> %a, <8 x i16> %b)
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI4_0)
 ; CHECK-NEXT:    vld $vr2, $a0, %pc_lo12(.LCPI4_0)
-; CHECK-NEXT:    vshuf.h $vr2, $vr1, $vr0
+; CHECK-NEXT:    vshuf.d $vr2, $vr1, $vr0
 ; CHECK-NEXT:    vori.b $vr0, $vr2, 0
 ; CHECK-NEXT:    ret
     %r = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 12, i32 13, i32 14, i32 15>
@@ -63,7 +66,7 @@ define <4 x i32> @widen_shuffle_mask_v4i32_to_v2i64(<4 x i32> %a, <4 x i32> %b)
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI5_0)
 ; CHECK-NEXT:    vld $vr2, $a0, %pc_lo12(.LCPI5_0)
-; CHECK-NEXT:    vshuf.w $vr2, $vr1, $vr0
+; CHECK-NEXT:    vshuf.d $vr2, $vr1, $vr0
 ; CHECK-NEXT:    vori.b $vr0, $vr2, 0
 ; CHECK-NEXT:    ret
     %r = shufflevector <4 x i32> %a, <4 x i32> %b, <4 x i32> <i32 0, i32 1, i32 6, i32 7>
@@ -73,9 +76,7 @@ define <4 x i32> @widen_shuffle_mask_v4i32_to_v2i64(<4 x i32> %a, <4 x i32> %b)
 define <16 x i8> @widen_shuffle_mask_v16i8_to_vpackev_h(<16 x i8> %a, <16 x i8> %b) {
 ; CHECK-LABEL: widen_shuffle_mask_v16i8_to_vpackev_h:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI6_0)
-; CHECK-NEXT:    vld $vr2, $a0, %pc_lo12(.LCPI6_0)
-; CHECK-NEXT:    vshuf.b $vr0, $vr1, $vr0, $vr2
+; CHECK-NEXT:    vpackev.h $vr0, $vr1, $vr0
 ; CHECK-NEXT:    ret
     %r = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 0, i32 1, i32 16, i32 17, i32 4, i32 5, i32 20, i32 21, i32 8, i32 9, i32 24, i32 25, i32 12, i32 13, i32 28, i32 29>
     ret <16 x i8> %r
@@ -84,9 +85,7 @@ define <16 x i8> @widen_shuffle_mask_v16i8_to_vpackev_h(<16 x i8> %a, <16 x i8>
 define <16 x i8> @widen_shuffle_mask_v16i8_to_vpackod_h(<16 x i8> %a, <16 x i8> %b) {
 ; CHECK-LABEL: widen_shuffle_mask_v16i8_to_vpackod_h:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI7_0)
-; CHECK-NEXT:    vld $vr2, $a0, %pc_lo12(.LCPI7_0)
-; CHECK-NEXT:    vshuf.b $vr0, $vr1, $vr0, $vr2
+; CHECK-NEXT:    vpackod.h $vr0, $vr1, $vr0
 ; CHECK-NEXT:    ret
     %r = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 2, i32 3, i32 18, i32 19, i32 6, i32 7, i32 22, i32 23, i32 10, i32 11, i32 26, i32 27, i32 14, i32 15, i32 30, i32 31>
     ret <16 x i8> %r
@@ -95,9 +94,7 @@ define <16 x i8> @widen_shuffle_mask_v16i8_to_vpackod_h(<16 x i8> %a, <16 x i8>
 define <16 x i8> @widen_shuffle_mask_v16i8_to_vpickev_h(<16 x i8> %a, <16 x i8> %b) {
 ; CHECK-LABEL: widen_shuffle_mask_v16i8_to_vpickev_h:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI8_0)
-; CHECK-NEXT:    vld $vr2, $a0, %pc_lo12(.LCPI8_0)
-; CHECK-NEXT:    vshuf.b $vr0, $vr1, $vr0, $vr2
+; CHECK-NEXT:    vpickev.h $vr0, $vr1, $vr0
 ; CHECK-NEXT:    ret
     %r = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 0, i32 1, i32 4, i32 5, i32 8, i32 9, i32 12, i32 13, i32 16, i32 17, i32 20, i32 21, i32 24, i32 25, i32 28, i32 29>
     ret <16 x i8> %r
@@ -106,9 +103,7 @@ define <16 x i8> @widen_shuffle_mask_v16i8_to_vpickev_h(<16 x i8> %a, <16 x i8>
 define <16 x i8> @widen_shuffle_mask_v16i8_to_vpickod_h(<16 x i8> %a, <16 x i8> %b) {
 ; CHECK-LABEL: widen_shuffle_mask_v16i8_to_vpickod_h:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI9_0)
-; CHECK-NEXT:    vld $vr2, $a0, %pc_lo12(.LCPI9_0)
-; CHECK-NEXT:    vshuf.b $vr0, $vr1, $vr0, $vr2
+; CHECK-NEXT:    vpickod.h $vr0, $vr1, $vr0
 ; CHECK-NEXT:    ret
     %r = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 2, i32 3, i32 6, i32 7, i32 10, i32 11, i32 14, i32 15, i32 18, i32 19, i32 22, i32 23, i32 26, i32 27, i32 30, i32 31>
     ret <16 x i8> %r
@@ -117,9 +112,7 @@ define <16 x i8> @widen_shuffle_mask_v16i8_to_vpickod_h(<16 x i8> %a, <16 x i8>
 define <16 x i8> @widen_shuffle_mask_v16i8_to_vilvl_h(<16 x i8> %a, <16 x i8> %b) {
 ; CHECK-LABEL: widen_shuffle_mask_v16i8_to_vilvl_h:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI10_0)
-; CHECK-NEXT:    vld $vr2, $a0, %pc_lo12(.LCPI10_0)
-; CHECK-NEXT:    vshuf.b $vr0, $vr1, $vr0, $vr2
+; CHECK-NEXT:    vilvl.h $vr0, $vr1, $vr0
 ; CHECK-NEXT:    ret
     %r = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 0, i32 1, i32 16, i32 17, i32 2, i32 3, i32 18, i32 19, i32 4, i32 5, i32 20, i32 21, i32 6, i32 7, i32 22, i32 23>
     ret <16 x i8> %r
@@ -128,9 +121,7 @@ define <16 x i8> @widen_shuffle_mask_v16i8_to_vilvl_h(<16 x i8> %a, <16 x i8> %b
 define <16 x i8> @widen_shuffle_mask_v16i8_to_vilvh_h(<16 x i8> %a, <16 x i8> %b) {
 ; CHECK-LABEL: widen_shuffle_mask_v16i8_to_vilvh_h:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    pcalau12i $a0, %pc_hi20(.LCPI11_0)
-; CHECK-NEXT:    vld $vr2, $a0, %pc_lo12(.LCPI11_0)
-; CHECK-NEXT:    vshuf.b $vr0, $vr1, $vr0, $vr2
+; CHECK-NEXT:    vilvh.h $vr0, $vr1, $vr0
 ; CHECK-NEXT:    ret
     %r = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 8, i32 9, i32 24, i32 25, i32 10, i32 11, i32 26, i32 27, i32 12, i32 13, i32 28, i32 29, i32 14, i32 15, i32 30, i32 31>
     ret <16 x i8> %r