Update aosp/master LLVM for rebase to r256229

[android-x86/external-llvm.git] / test / CodeGen / AArch64 / arm64-vabs.ll

diff --git a/test/CodeGen/AArch64/arm64-vabs.ll b/test/CodeGen/AArch64/arm64-vabs.ll
index fae2b90..c180008 100644 (file)
--- a/test/CodeGen/AArch64/arm64-vabs.ll
+++ b/test/CodeGen/AArch64/arm64-vabs.ll
@@ -4,8 +4,8 @@
 define <8 x i16> @sabdl8h(<8 x i8>* %A, <8 x i8>* %B) nounwind {
 ;CHECK-LABEL: sabdl8h:
 ;CHECK: sabdl.8h
-        %tmp1 = load <8 x i8>* %A
-        %tmp2 = load <8 x i8>* %B
+        %tmp1 = load <8 x i8>, <8 x i8>* %A
+        %tmp2 = load <8 x i8>, <8 x i8>* %B
         %tmp3 = call <8 x i8> @llvm.aarch64.neon.sabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
         %tmp4 = zext <8 x i8> %tmp3 to <8 x i16>
         ret <8 x i16> %tmp4
@@ -14,8 +14,8 @@ define <8 x i16> @sabdl8h(<8 x i8>* %A, <8 x i8>* %B) nounwind {
 define <4 x i32> @sabdl4s(<4 x i16>* %A, <4 x i16>* %B) nounwind {
 ;CHECK-LABEL: sabdl4s:
 ;CHECK: sabdl.4s
-        %tmp1 = load <4 x i16>* %A
-        %tmp2 = load <4 x i16>* %B
+        %tmp1 = load <4 x i16>, <4 x i16>* %A
+        %tmp2 = load <4 x i16>, <4 x i16>* %B
         %tmp3 = call <4 x i16> @llvm.aarch64.neon.sabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
         %tmp4 = zext <4 x i16> %tmp3 to <4 x i32>
         ret <4 x i32> %tmp4
@@ -24,8 +24,8 @@ define <4 x i32> @sabdl4s(<4 x i16>* %A, <4 x i16>* %B) nounwind {
 define <2 x i64> @sabdl2d(<2 x i32>* %A, <2 x i32>* %B) nounwind {
 ;CHECK-LABEL: sabdl2d:
 ;CHECK: sabdl.2d
-        %tmp1 = load <2 x i32>* %A
-        %tmp2 = load <2 x i32>* %B
+        %tmp1 = load <2 x i32>, <2 x i32>* %A
+        %tmp2 = load <2 x i32>, <2 x i32>* %B
         %tmp3 = call <2 x i32> @llvm.aarch64.neon.sabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
         %tmp4 = zext <2 x i32> %tmp3 to <2 x i64>
         ret <2 x i64> %tmp4
@@ -34,8 +34,8 @@ define <2 x i64> @sabdl2d(<2 x i32>* %A, <2 x i32>* %B) nounwind {
 define <8 x i16> @sabdl2_8h(<16 x i8>* %A, <16 x i8>* %B) nounwind {
 ;CHECK-LABEL: sabdl2_8h:
 ;CHECK: sabdl2.8h
-        %load1 = load <16 x i8>* %A
-        %load2 = load <16 x i8>* %B
+        %load1 = load <16 x i8>, <16 x i8>* %A
+        %load2 = load <16 x i8>, <16 x i8>* %B
         %tmp1 = shufflevector <16 x i8> %load1, <16 x i8> undef, <8 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
         %tmp2 = shufflevector <16 x i8> %load2, <16 x i8> undef, <8 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
         %tmp3 = call <8 x i8> @llvm.aarch64.neon.sabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
@@ -46,8 +46,8 @@ define <8 x i16> @sabdl2_8h(<16 x i8>* %A, <16 x i8>* %B) nounwind {
 define <4 x i32> @sabdl2_4s(<8 x i16>* %A, <8 x i16>* %B) nounwind {
 ;CHECK-LABEL: sabdl2_4s:
 ;CHECK: sabdl2.4s
-        %load1 = load <8 x i16>* %A
-        %load2 = load <8 x i16>* %B
+        %load1 = load <8 x i16>, <8 x i16>* %A
+        %load2 = load <8 x i16>, <8 x i16>* %B
         %tmp1 = shufflevector <8 x i16> %load1, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
         %tmp2 = shufflevector <8 x i16> %load2, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
         %tmp3 = call <4 x i16> @llvm.aarch64.neon.sabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
@@ -58,8 +58,8 @@ define <4 x i32> @sabdl2_4s(<8 x i16>* %A, <8 x i16>* %B) nounwind {
 define <2 x i64> @sabdl2_2d(<4 x i32>* %A, <4 x i32>* %B) nounwind {
 ;CHECK-LABEL: sabdl2_2d:
 ;CHECK: sabdl2.2d
-        %load1 = load <4 x i32>* %A
-        %load2 = load <4 x i32>* %B
+        %load1 = load <4 x i32>, <4 x i32>* %A
+        %load2 = load <4 x i32>, <4 x i32>* %B
         %tmp1 = shufflevector <4 x i32> %load1, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
         %tmp2 = shufflevector <4 x i32> %load2, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
         %tmp3 = call <2 x i32> @llvm.aarch64.neon.sabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
@@ -70,8 +70,8 @@ define <2 x i64> @sabdl2_2d(<4 x i32>* %A, <4 x i32>* %B) nounwind {
 define <8 x i16> @uabdl8h(<8 x i8>* %A, <8 x i8>* %B) nounwind {
 ;CHECK-LABEL: uabdl8h:
 ;CHECK: uabdl.8h
-  %tmp1 = load <8 x i8>* %A
-  %tmp2 = load <8 x i8>* %B
+  %tmp1 = load <8 x i8>, <8 x i8>* %A
+  %tmp2 = load <8 x i8>, <8 x i8>* %B
   %tmp3 = call <8 x i8> @llvm.aarch64.neon.uabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
   %tmp4 = zext <8 x i8> %tmp3 to <8 x i16>
   ret <8 x i16> %tmp4
@@ -80,8 +80,8 @@ define <8 x i16> @uabdl8h(<8 x i8>* %A, <8 x i8>* %B) nounwind {
 define <4 x i32> @uabdl4s(<4 x i16>* %A, <4 x i16>* %B) nounwind {
 ;CHECK-LABEL: uabdl4s:
 ;CHECK: uabdl.4s
-  %tmp1 = load <4 x i16>* %A
-  %tmp2 = load <4 x i16>* %B
+  %tmp1 = load <4 x i16>, <4 x i16>* %A
+  %tmp2 = load <4 x i16>, <4 x i16>* %B
   %tmp3 = call <4 x i16> @llvm.aarch64.neon.uabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
   %tmp4 = zext <4 x i16> %tmp3 to <4 x i32>
   ret <4 x i32> %tmp4
@@ -90,8 +90,8 @@ define <4 x i32> @uabdl4s(<4 x i16>* %A, <4 x i16>* %B) nounwind {
 define <2 x i64> @uabdl2d(<2 x i32>* %A, <2 x i32>* %B) nounwind {
 ;CHECK-LABEL: uabdl2d:
 ;CHECK: uabdl.2d
-  %tmp1 = load <2 x i32>* %A
-  %tmp2 = load <2 x i32>* %B
+  %tmp1 = load <2 x i32>, <2 x i32>* %A
+  %tmp2 = load <2 x i32>, <2 x i32>* %B
   %tmp3 = call <2 x i32> @llvm.aarch64.neon.uabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
   %tmp4 = zext <2 x i32> %tmp3 to <2 x i64>
   ret <2 x i64> %tmp4
@@ -100,8 +100,8 @@ define <2 x i64> @uabdl2d(<2 x i32>* %A, <2 x i32>* %B) nounwind {
 define <8 x i16> @uabdl2_8h(<16 x i8>* %A, <16 x i8>* %B) nounwind {
 ;CHECK-LABEL: uabdl2_8h:
 ;CHECK: uabdl2.8h
-  %load1 = load <16 x i8>* %A
-  %load2 = load <16 x i8>* %B
+  %load1 = load <16 x i8>, <16 x i8>* %A
+  %load2 = load <16 x i8>, <16 x i8>* %B
   %tmp1 = shufflevector <16 x i8> %load1, <16 x i8> undef, <8 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
   %tmp2 = shufflevector <16 x i8> %load2, <16 x i8> undef, <8 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
 
@@ -113,8 +113,8 @@ define <8 x i16> @uabdl2_8h(<16 x i8>* %A, <16 x i8>* %B) nounwind {
 define <4 x i32> @uabdl2_4s(<8 x i16>* %A, <8 x i16>* %B) nounwind {
 ;CHECK-LABEL: uabdl2_4s:
 ;CHECK: uabdl2.4s
-  %load1 = load <8 x i16>* %A
-  %load2 = load <8 x i16>* %B
+  %load1 = load <8 x i16>, <8 x i16>* %A
+  %load2 = load <8 x i16>, <8 x i16>* %B
   %tmp1 = shufflevector <8 x i16> %load1, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
   %tmp2 = shufflevector <8 x i16> %load2, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
   %tmp3 = call <4 x i16> @llvm.aarch64.neon.uabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
@@ -125,8 +125,8 @@ define <4 x i32> @uabdl2_4s(<8 x i16>* %A, <8 x i16>* %B) nounwind {
 define <2 x i64> @uabdl2_2d(<4 x i32>* %A, <4 x i32>* %B) nounwind {
 ;CHECK-LABEL: uabdl2_2d:
 ;CHECK: uabdl2.2d
-  %load1 = load <4 x i32>* %A
-  %load2 = load <4 x i32>* %B
+  %load1 = load <4 x i32>, <4 x i32>* %A
+  %load2 = load <4 x i32>, <4 x i32>* %B
   %tmp1 = shufflevector <4 x i32> %load1, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
   %tmp2 = shufflevector <4 x i32> %load2, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
   %tmp3 = call <2 x i32> @llvm.aarch64.neon.uabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
@@ -134,11 +134,77 @@ define <2 x i64> @uabdl2_2d(<4 x i32>* %A, <4 x i32>* %B) nounwind {
   ret <2 x i64> %tmp4
 }
 
+define i16 @uabdl8h_log2_shuffle(<16 x i8>* %a, <16 x i8>* %b) {
+; CHECK-LABEL: uabdl8h_log2_shuffle
+; CHECK: uabdl2.8h
+; CHECK: uabdl.8h
+  %aload = load <16 x i8>, <16 x i8>* %a, align 1
+  %bload = load <16 x i8>, <16 x i8>* %b, align 1
+  %aext = zext <16 x i8> %aload to <16 x i16>
+  %bext = zext <16 x i8> %bload to <16 x i16>
+  %abdiff = sub nsw <16 x i16> %aext, %bext
+  %abcmp = icmp slt <16 x i16> %abdiff, zeroinitializer
+  %ababs = sub nsw <16 x i16> zeroinitializer, %abdiff
+  %absel = select <16 x i1> %abcmp, <16 x i16> %ababs, <16 x i16> %abdiff
+  %rdx.shuf = shufflevector <16 x i16> %absel, <16 x i16> undef, <16 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+  %bin1.rdx = add <16 x i16> %absel, %rdx.shuf
+  %rdx.shufx = shufflevector <16 x i16> %bin1.rdx, <16 x i16> undef, <16 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+  %bin.rdx = add <16 x i16> %bin1.rdx, %rdx.shufx
+  %rdx.shuf136 = shufflevector <16 x i16> %bin.rdx, <16 x i16> undef, <16 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+  %bin.rdx137 = add <16 x i16> %bin.rdx, %rdx.shuf136
+  %rdx.shuf138 = shufflevector <16 x i16> %bin.rdx137, <16 x i16> undef, <16 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+  %bin.rdx139 = add <16 x i16> %bin.rdx137, %rdx.shuf138
+  %reduced_v = extractelement <16 x i16> %bin.rdx139, i16 0
+  ret i16 %reduced_v
+}
+
+define i32 @uabdl4s_log2_shuffle(<8 x i16>* %a, <8 x i16>* %b) {
+; CHECK-LABEL: uabdl4s_log2_shuffle
+; CHECK: uabdl2.4s
+; CHECK: uabdl.4s
+  %aload = load <8 x i16>, <8 x i16>* %a, align 1
+  %bload = load <8 x i16>, <8 x i16>* %b, align 1
+  %aext = zext <8 x i16> %aload to <8 x i32>
+  %bext = zext <8 x i16> %bload to <8 x i32>
+  %abdiff = sub nsw <8 x i32> %aext, %bext
+  %abcmp = icmp slt <8 x i32> %abdiff, zeroinitializer
+  %ababs = sub nsw <8 x i32> zeroinitializer, %abdiff
+  %absel = select <8 x i1> %abcmp, <8 x i32> %ababs, <8 x i32> %abdiff
+  %rdx.shuf = shufflevector <8 x i32> %absel, <8 x i32> undef, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef>
+  %bin.rdx = add <8 x i32> %absel, %rdx.shuf
+  %rdx.shuf136 = shufflevector <8 x i32> %bin.rdx, <8 x i32> undef, <8 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+  %bin.rdx137 = add <8 x i32> %bin.rdx, %rdx.shuf136
+  %rdx.shuf138 = shufflevector <8 x i32> %bin.rdx137, <8 x i32> undef, <8 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+  %bin.rdx139 = add <8 x i32> %bin.rdx137, %rdx.shuf138
+  %reduced_v = extractelement <8 x i32> %bin.rdx139, i32 0
+  ret i32 %reduced_v
+}
+
+define i64 @uabdl2d_log2_shuffle(<4 x i32>* %a, <4 x i32>* %b, i32 %h) {
+; CHECK: uabdl2d_log2_shuffle
+; CHECK: uabdl2.2d
+; CHECK: uabdl.2d
+  %aload = load <4 x i32>, <4 x i32>* %a, align 1
+  %bload = load <4 x i32>, <4 x i32>* %b, align 1
+  %aext = zext <4 x i32> %aload to <4 x i64>
+  %bext = zext <4 x i32> %bload to <4 x i64>
+  %abdiff = sub nsw <4 x i64> %aext, %bext
+  %abcmp = icmp slt <4 x i64> %abdiff, zeroinitializer
+  %ababs = sub nsw <4 x i64> zeroinitializer, %abdiff
+  %absel = select <4 x i1> %abcmp, <4 x i64> %ababs, <4 x i64> %abdiff
+  %rdx.shuf136 = shufflevector <4 x i64> %absel, <4 x i64> undef, <4 x i32> <i32 2, i32 3, i32 undef, i32 undef>
+  %bin.rdx137 = add <4 x i64> %absel, %rdx.shuf136
+  %rdx.shuf138 = shufflevector <4 x i64> %bin.rdx137, <4 x i64> undef, <4 x i32> <i32 1, i32 undef, i32 undef, i32 undef>
+  %bin.rdx139 = add <4 x i64> %bin.rdx137, %rdx.shuf138
+  %reduced_v = extractelement <4 x i64> %bin.rdx139, i16 0
+  ret i64 %reduced_v
+}
+
 define <2 x float> @fabd_2s(<2 x float>* %A, <2 x float>* %B) nounwind {
 ;CHECK-LABEL: fabd_2s:
 ;CHECK: fabd.2s
-        %tmp1 = load <2 x float>* %A
-        %tmp2 = load <2 x float>* %B
+        %tmp1 = load <2 x float>, <2 x float>* %A
+        %tmp2 = load <2 x float>, <2 x float>* %B
         %tmp3 = call <2 x float> @llvm.aarch64.neon.fabd.v2f32(<2 x float> %tmp1, <2 x float> %tmp2)
         ret <2 x float> %tmp3
 }
@@ -146,8 +212,8 @@ define <2 x float> @fabd_2s(<2 x float>* %A, <2 x float>* %B) nounwind {
 define <4 x float> @fabd_4s(<4 x float>* %A, <4 x float>* %B) nounwind {
 ;CHECK-LABEL: fabd_4s:
 ;CHECK: fabd.4s
-        %tmp1 = load <4 x float>* %A
-        %tmp2 = load <4 x float>* %B
+        %tmp1 = load <4 x float>, <4 x float>* %A
+        %tmp2 = load <4 x float>, <4 x float>* %B
         %tmp3 = call <4 x float> @llvm.aarch64.neon.fabd.v4f32(<4 x float> %tmp1, <4 x float> %tmp2)
         ret <4 x float> %tmp3
 }
@@ -155,8 +221,8 @@ define <4 x float> @fabd_4s(<4 x float>* %A, <4 x float>* %B) nounwind {
 define <2 x double> @fabd_2d(<2 x double>* %A, <2 x double>* %B) nounwind {
 ;CHECK-LABEL: fabd_2d:
 ;CHECK: fabd.2d
-        %tmp1 = load <2 x double>* %A
-        %tmp2 = load <2 x double>* %B
+        %tmp1 = load <2 x double>, <2 x double>* %A
+        %tmp2 = load <2 x double>, <2 x double>* %B
         %tmp3 = call <2 x double> @llvm.aarch64.neon.fabd.v2f64(<2 x double> %tmp1, <2 x double> %tmp2)
         ret <2 x double> %tmp3
 }
@@ -168,8 +234,8 @@ declare <2 x double> @llvm.aarch64.neon.fabd.v2f64(<2 x double>, <2 x double>) n
 define <8 x i8> @sabd_8b(<8 x i8>* %A, <8 x i8>* %B) nounwind {
 ;CHECK-LABEL: sabd_8b:
 ;CHECK: sabd.8b
-        %tmp1 = load <8 x i8>* %A
-        %tmp2 = load <8 x i8>* %B
+        %tmp1 = load <8 x i8>, <8 x i8>* %A
+        %tmp2 = load <8 x i8>, <8 x i8>* %B
         %tmp3 = call <8 x i8> @llvm.aarch64.neon.sabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
         ret <8 x i8> %tmp3
 }
@@ -177,8 +243,8 @@ define <8 x i8> @sabd_8b(<8 x i8>* %A, <8 x i8>* %B) nounwind {
 define <16 x i8> @sabd_16b(<16 x i8>* %A, <16 x i8>* %B) nounwind {
 ;CHECK-LABEL: sabd_16b:
 ;CHECK: sabd.16b
-        %tmp1 = load <16 x i8>* %A
-        %tmp2 = load <16 x i8>* %B
+        %tmp1 = load <16 x i8>, <16 x i8>* %A
+        %tmp2 = load <16 x i8>, <16 x i8>* %B
         %tmp3 = call <16 x i8> @llvm.aarch64.neon.sabd.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
         ret <16 x i8> %tmp3
 }
@@ -186,8 +252,8 @@ define <16 x i8> @sabd_16b(<16 x i8>* %A, <16 x i8>* %B) nounwind {
 define <4 x i16> @sabd_4h(<4 x i16>* %A, <4 x i16>* %B) nounwind {
 ;CHECK-LABEL: sabd_4h:
 ;CHECK: sabd.4h
-        %tmp1 = load <4 x i16>* %A
-        %tmp2 = load <4 x i16>* %B
+        %tmp1 = load <4 x i16>, <4 x i16>* %A
+        %tmp2 = load <4 x i16>, <4 x i16>* %B
         %tmp3 = call <4 x i16> @llvm.aarch64.neon.sabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
         ret <4 x i16> %tmp3
 }
@@ -195,8 +261,8 @@ define <4 x i16> @sabd_4h(<4 x i16>* %A, <4 x i16>* %B) nounwind {
 define <8 x i16> @sabd_8h(<8 x i16>* %A, <8 x i16>* %B) nounwind {
 ;CHECK-LABEL: sabd_8h:
 ;CHECK: sabd.8h
-        %tmp1 = load <8 x i16>* %A
-        %tmp2 = load <8 x i16>* %B
+        %tmp1 = load <8 x i16>, <8 x i16>* %A
+        %tmp2 = load <8 x i16>, <8 x i16>* %B
         %tmp3 = call <8 x i16> @llvm.aarch64.neon.sabd.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
         ret <8 x i16> %tmp3
 }
@@ -204,8 +270,8 @@ define <8 x i16> @sabd_8h(<8 x i16>* %A, <8 x i16>* %B) nounwind {
 define <2 x i32> @sabd_2s(<2 x i32>* %A, <2 x i32>* %B) nounwind {
 ;CHECK-LABEL: sabd_2s:
 ;CHECK: sabd.2s
-        %tmp1 = load <2 x i32>* %A
-        %tmp2 = load <2 x i32>* %B
+        %tmp1 = load <2 x i32>, <2 x i32>* %A
+        %tmp2 = load <2 x i32>, <2 x i32>* %B
         %tmp3 = call <2 x i32> @llvm.aarch64.neon.sabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
         ret <2 x i32> %tmp3
 }
@@ -213,8 +279,8 @@ define <2 x i32> @sabd_2s(<2 x i32>* %A, <2 x i32>* %B) nounwind {
 define <4 x i32> @sabd_4s(<4 x i32>* %A, <4 x i32>* %B) nounwind {
 ;CHECK-LABEL: sabd_4s:
 ;CHECK: sabd.4s
-        %tmp1 = load <4 x i32>* %A
-        %tmp2 = load <4 x i32>* %B
+        %tmp1 = load <4 x i32>, <4 x i32>* %A
+        %tmp2 = load <4 x i32>, <4 x i32>* %B
         %tmp3 = call <4 x i32> @llvm.aarch64.neon.sabd.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
         ret <4 x i32> %tmp3
 }
@@ -229,8 +295,8 @@ declare <4 x i32> @llvm.aarch64.neon.sabd.v4i32(<4 x i32>, <4 x i32>) nounwind r
 define <8 x i8> @uabd_8b(<8 x i8>* %A, <8 x i8>* %B) nounwind {
 ;CHECK-LABEL: uabd_8b:
 ;CHECK: uabd.8b
-        %tmp1 = load <8 x i8>* %A
-        %tmp2 = load <8 x i8>* %B
+        %tmp1 = load <8 x i8>, <8 x i8>* %A
+        %tmp2 = load <8 x i8>, <8 x i8>* %B
         %tmp3 = call <8 x i8> @llvm.aarch64.neon.uabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
         ret <8 x i8> %tmp3
 }
@@ -238,8 +304,8 @@ define <8 x i8> @uabd_8b(<8 x i8>* %A, <8 x i8>* %B) nounwind {
 define <16 x i8> @uabd_16b(<16 x i8>* %A, <16 x i8>* %B) nounwind {
 ;CHECK-LABEL: uabd_16b:
 ;CHECK: uabd.16b
-        %tmp1 = load <16 x i8>* %A
-        %tmp2 = load <16 x i8>* %B
+        %tmp1 = load <16 x i8>, <16 x i8>* %A
+        %tmp2 = load <16 x i8>, <16 x i8>* %B
         %tmp3 = call <16 x i8> @llvm.aarch64.neon.uabd.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
         ret <16 x i8> %tmp3
 }
@@ -247,8 +313,8 @@ define <16 x i8> @uabd_16b(<16 x i8>* %A, <16 x i8>* %B) nounwind {
 define <4 x i16> @uabd_4h(<4 x i16>* %A, <4 x i16>* %B) nounwind {
 ;CHECK-LABEL: uabd_4h:
 ;CHECK: uabd.4h
-        %tmp1 = load <4 x i16>* %A
-        %tmp2 = load <4 x i16>* %B
+        %tmp1 = load <4 x i16>, <4 x i16>* %A
+        %tmp2 = load <4 x i16>, <4 x i16>* %B
         %tmp3 = call <4 x i16> @llvm.aarch64.neon.uabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
         ret <4 x i16> %tmp3
 }
@@ -256,8 +322,8 @@ define <4 x i16> @uabd_4h(<4 x i16>* %A, <4 x i16>* %B) nounwind {
 define <8 x i16> @uabd_8h(<8 x i16>* %A, <8 x i16>* %B) nounwind {
 ;CHECK-LABEL: uabd_8h:
 ;CHECK: uabd.8h
-        %tmp1 = load <8 x i16>* %A
-        %tmp2 = load <8 x i16>* %B
+        %tmp1 = load <8 x i16>, <8 x i16>* %A
+        %tmp2 = load <8 x i16>, <8 x i16>* %B
         %tmp3 = call <8 x i16> @llvm.aarch64.neon.uabd.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
         ret <8 x i16> %tmp3
 }
@@ -265,8 +331,8 @@ define <8 x i16> @uabd_8h(<8 x i16>* %A, <8 x i16>* %B) nounwind {
 define <2 x i32> @uabd_2s(<2 x i32>* %A, <2 x i32>* %B) nounwind {
 ;CHECK-LABEL: uabd_2s:
 ;CHECK: uabd.2s
-        %tmp1 = load <2 x i32>* %A
-        %tmp2 = load <2 x i32>* %B
+        %tmp1 = load <2 x i32>, <2 x i32>* %A
+        %tmp2 = load <2 x i32>, <2 x i32>* %B
         %tmp3 = call <2 x i32> @llvm.aarch64.neon.uabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
         ret <2 x i32> %tmp3
 }
@@ -274,8 +340,8 @@ define <2 x i32> @uabd_2s(<2 x i32>* %A, <2 x i32>* %B) nounwind {
 define <4 x i32> @uabd_4s(<4 x i32>* %A, <4 x i32>* %B) nounwind {
 ;CHECK-LABEL: uabd_4s:
 ;CHECK: uabd.4s
-        %tmp1 = load <4 x i32>* %A
-        %tmp2 = load <4 x i32>* %B
+        %tmp1 = load <4 x i32>, <4 x i32>* %A
+        %tmp2 = load <4 x i32>, <4 x i32>* %B
         %tmp3 = call <4 x i32> @llvm.aarch64.neon.uabd.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
         ret <4 x i32> %tmp3
 }
@@ -290,7 +356,7 @@ declare <4 x i32> @llvm.aarch64.neon.uabd.v4i32(<4 x i32>, <4 x i32>) nounwind r
 define <8 x i8> @sqabs_8b(<8 x i8>* %A) nounwind {
 ;CHECK-LABEL: sqabs_8b:
 ;CHECK: sqabs.8b
-        %tmp1 = load <8 x i8>* %A
+        %tmp1 = load <8 x i8>, <8 x i8>* %A
         %tmp3 = call <8 x i8> @llvm.aarch64.neon.sqabs.v8i8(<8 x i8> %tmp1)
         ret <8 x i8> %tmp3
 }
@@ -298,7 +364,7 @@ define <8 x i8> @sqabs_8b(<8 x i8>* %A) nounwind {
 define <16 x i8> @sqabs_16b(<16 x i8>* %A) nounwind {
 ;CHECK-LABEL: sqabs_16b:
 ;CHECK: sqabs.16b
-        %tmp1 = load <16 x i8>* %A
+        %tmp1 = load <16 x i8>, <16 x i8>* %A
         %tmp3 = call <16 x i8> @llvm.aarch64.neon.sqabs.v16i8(<16 x i8> %tmp1)
         ret <16 x i8> %tmp3
 }
@@ -306,7 +372,7 @@ define <16 x i8> @sqabs_16b(<16 x i8>* %A) nounwind {
 define <4 x i16> @sqabs_4h(<4 x i16>* %A) nounwind {
 ;CHECK-LABEL: sqabs_4h:
 ;CHECK: sqabs.4h
-        %tmp1 = load <4 x i16>* %A
+        %tmp1 = load <4 x i16>, <4 x i16>* %A
         %tmp3 = call <4 x i16> @llvm.aarch64.neon.sqabs.v4i16(<4 x i16> %tmp1)
         ret <4 x i16> %tmp3
 }
@@ -314,7 +380,7 @@ define <4 x i16> @sqabs_4h(<4 x i16>* %A) nounwind {
 define <8 x i16> @sqabs_8h(<8 x i16>* %A) nounwind {
 ;CHECK-LABEL: sqabs_8h:
 ;CHECK: sqabs.8h
-        %tmp1 = load <8 x i16>* %A
+        %tmp1 = load <8 x i16>, <8 x i16>* %A
         %tmp3 = call <8 x i16> @llvm.aarch64.neon.sqabs.v8i16(<8 x i16> %tmp1)
         ret <8 x i16> %tmp3
 }
@@ -322,7 +388,7 @@ define <8 x i16> @sqabs_8h(<8 x i16>* %A) nounwind {
 define <2 x i32> @sqabs_2s(<2 x i32>* %A) nounwind {
 ;CHECK-LABEL: sqabs_2s:
 ;CHECK: sqabs.2s
-        %tmp1 = load <2 x i32>* %A
+        %tmp1 = load <2 x i32>, <2 x i32>* %A
         %tmp3 = call <2 x i32> @llvm.aarch64.neon.sqabs.v2i32(<2 x i32> %tmp1)
         ret <2 x i32> %tmp3
 }
@@ -330,7 +396,7 @@ define <2 x i32> @sqabs_2s(<2 x i32>* %A) nounwind {
 define <4 x i32> @sqabs_4s(<4 x i32>* %A) nounwind {
 ;CHECK-LABEL: sqabs_4s:
 ;CHECK: sqabs.4s
-        %tmp1 = load <4 x i32>* %A
+        %tmp1 = load <4 x i32>, <4 x i32>* %A
         %tmp3 = call <4 x i32> @llvm.aarch64.neon.sqabs.v4i32(<4 x i32> %tmp1)
         ret <4 x i32> %tmp3
 }
@@ -345,7 +411,7 @@ declare <4 x i32> @llvm.aarch64.neon.sqabs.v4i32(<4 x i32>) nounwind readnone
 define <8 x i8> @sqneg_8b(<8 x i8>* %A) nounwind {
 ;CHECK-LABEL: sqneg_8b:
 ;CHECK: sqneg.8b
-        %tmp1 = load <8 x i8>* %A
+        %tmp1 = load <8 x i8>, <8 x i8>* %A
         %tmp3 = call <8 x i8> @llvm.aarch64.neon.sqneg.v8i8(<8 x i8> %tmp1)
         ret <8 x i8> %tmp3
 }
@@ -353,7 +419,7 @@ define <8 x i8> @sqneg_8b(<8 x i8>* %A) nounwind {
 define <16 x i8> @sqneg_16b(<16 x i8>* %A) nounwind {
 ;CHECK-LABEL: sqneg_16b:
 ;CHECK: sqneg.16b
-        %tmp1 = load <16 x i8>* %A
+        %tmp1 = load <16 x i8>, <16 x i8>* %A
         %tmp3 = call <16 x i8> @llvm.aarch64.neon.sqneg.v16i8(<16 x i8> %tmp1)
         ret <16 x i8> %tmp3
 }
@@ -361,7 +427,7 @@ define <16 x i8> @sqneg_16b(<16 x i8>* %A) nounwind {
 define <4 x i16> @sqneg_4h(<4 x i16>* %A) nounwind {
 ;CHECK-LABEL: sqneg_4h:
 ;CHECK: sqneg.4h
-        %tmp1 = load <4 x i16>* %A
+        %tmp1 = load <4 x i16>, <4 x i16>* %A
         %tmp3 = call <4 x i16> @llvm.aarch64.neon.sqneg.v4i16(<4 x i16> %tmp1)
         ret <4 x i16> %tmp3
 }
@@ -369,7 +435,7 @@ define <4 x i16> @sqneg_4h(<4 x i16>* %A) nounwind {
 define <8 x i16> @sqneg_8h(<8 x i16>* %A) nounwind {
 ;CHECK-LABEL: sqneg_8h:
 ;CHECK: sqneg.8h
-        %tmp1 = load <8 x i16>* %A
+        %tmp1 = load <8 x i16>, <8 x i16>* %A
         %tmp3 = call <8 x i16> @llvm.aarch64.neon.sqneg.v8i16(<8 x i16> %tmp1)
         ret <8 x i16> %tmp3
 }
@@ -377,7 +443,7 @@ define <8 x i16> @sqneg_8h(<8 x i16>* %A) nounwind {
 define <2 x i32> @sqneg_2s(<2 x i32>* %A) nounwind {
 ;CHECK-LABEL: sqneg_2s:
 ;CHECK: sqneg.2s
-        %tmp1 = load <2 x i32>* %A
+        %tmp1 = load <2 x i32>, <2 x i32>* %A
         %tmp3 = call <2 x i32> @llvm.aarch64.neon.sqneg.v2i32(<2 x i32> %tmp1)
         ret <2 x i32> %tmp3
 }
@@ -385,7 +451,7 @@ define <2 x i32> @sqneg_2s(<2 x i32>* %A) nounwind {
 define <4 x i32> @sqneg_4s(<4 x i32>* %A) nounwind {
 ;CHECK-LABEL: sqneg_4s:
 ;CHECK: sqneg.4s
-        %tmp1 = load <4 x i32>* %A
+        %tmp1 = load <4 x i32>, <4 x i32>* %A
         %tmp3 = call <4 x i32> @llvm.aarch64.neon.sqneg.v4i32(<4 x i32> %tmp1)
         ret <4 x i32> %tmp3
 }
@@ -400,7 +466,7 @@ declare <4 x i32> @llvm.aarch64.neon.sqneg.v4i32(<4 x i32>) nounwind readnone
 define <8 x i8> @abs_8b(<8 x i8>* %A) nounwind {
 ;CHECK-LABEL: abs_8b:
 ;CHECK: abs.8b
-        %tmp1 = load <8 x i8>* %A
+        %tmp1 = load <8 x i8>, <8 x i8>* %A
         %tmp3 = call <8 x i8> @llvm.aarch64.neon.abs.v8i8(<8 x i8> %tmp1)
         ret <8 x i8> %tmp3
 }
@@ -408,7 +474,7 @@ define <8 x i8> @abs_8b(<8 x i8>* %A) nounwind {
 define <16 x i8> @abs_16b(<16 x i8>* %A) nounwind {
 ;CHECK-LABEL: abs_16b:
 ;CHECK: abs.16b
-        %tmp1 = load <16 x i8>* %A
+        %tmp1 = load <16 x i8>, <16 x i8>* %A
         %tmp3 = call <16 x i8> @llvm.aarch64.neon.abs.v16i8(<16 x i8> %tmp1)
         ret <16 x i8> %tmp3
 }
@@ -416,7 +482,7 @@ define <16 x i8> @abs_16b(<16 x i8>* %A) nounwind {
 define <4 x i16> @abs_4h(<4 x i16>* %A) nounwind {
 ;CHECK-LABEL: abs_4h:
 ;CHECK: abs.4h
-        %tmp1 = load <4 x i16>* %A
+        %tmp1 = load <4 x i16>, <4 x i16>* %A
         %tmp3 = call <4 x i16> @llvm.aarch64.neon.abs.v4i16(<4 x i16> %tmp1)
         ret <4 x i16> %tmp3
 }
@@ -424,7 +490,7 @@ define <4 x i16> @abs_4h(<4 x i16>* %A) nounwind {
 define <8 x i16> @abs_8h(<8 x i16>* %A) nounwind {
 ;CHECK-LABEL: abs_8h:
 ;CHECK: abs.8h
-        %tmp1 = load <8 x i16>* %A
+        %tmp1 = load <8 x i16>, <8 x i16>* %A
         %tmp3 = call <8 x i16> @llvm.aarch64.neon.abs.v8i16(<8 x i16> %tmp1)
         ret <8 x i16> %tmp3
 }
@@ -432,7 +498,7 @@ define <8 x i16> @abs_8h(<8 x i16>* %A) nounwind {
 define <2 x i32> @abs_2s(<2 x i32>* %A) nounwind {
 ;CHECK-LABEL: abs_2s:
 ;CHECK: abs.2s
-        %tmp1 = load <2 x i32>* %A
+        %tmp1 = load <2 x i32>, <2 x i32>* %A
         %tmp3 = call <2 x i32> @llvm.aarch64.neon.abs.v2i32(<2 x i32> %tmp1)
         ret <2 x i32> %tmp3
 }
@@ -440,7 +506,7 @@ define <2 x i32> @abs_2s(<2 x i32>* %A) nounwind {
 define <4 x i32> @abs_4s(<4 x i32>* %A) nounwind {
 ;CHECK-LABEL: abs_4s:
 ;CHECK: abs.4s
-        %tmp1 = load <4 x i32>* %A
+        %tmp1 = load <4 x i32>, <4 x i32>* %A
         %tmp3 = call <4 x i32> @llvm.aarch64.neon.abs.v4i32(<4 x i32> %tmp1)
         ret <4 x i32> %tmp3
 }
@@ -471,9 +537,9 @@ declare i64 @llvm.aarch64.neon.abs.i64(i64) nounwind readnone
 define <8 x i16> @sabal8h(<8 x i8>* %A, <8 x i8>* %B,  <8 x i16>* %C) nounwind {
 ;CHECK-LABEL: sabal8h:
 ;CHECK: sabal.8h
-        %tmp1 = load <8 x i8>* %A
-        %tmp2 = load <8 x i8>* %B
-        %tmp3 = load <8 x i16>* %C
+        %tmp1 = load <8 x i8>, <8 x i8>* %A
+        %tmp2 = load <8 x i8>, <8 x i8>* %B
+        %tmp3 = load <8 x i16>, <8 x i16>* %C
         %tmp4 = call <8 x i8> @llvm.aarch64.neon.sabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
         %tmp4.1 = zext <8 x i8> %tmp4 to <8 x i16>
         %tmp5 = add <8 x i16> %tmp3, %tmp4.1
@@ -483,9 +549,9 @@ define <8 x i16> @sabal8h(<8 x i8>* %A, <8 x i8>* %B,  <8 x i16>* %C) nounwind {
 define <4 x i32> @sabal4s(<4 x i16>* %A, <4 x i16>* %B, <4 x i32>* %C) nounwind {
 ;CHECK-LABEL: sabal4s:
 ;CHECK: sabal.4s
-        %tmp1 = load <4 x i16>* %A
-        %tmp2 = load <4 x i16>* %B
-        %tmp3 = load <4 x i32>* %C
+        %tmp1 = load <4 x i16>, <4 x i16>* %A
+        %tmp2 = load <4 x i16>, <4 x i16>* %B
+        %tmp3 = load <4 x i32>, <4 x i32>* %C
         %tmp4 = call <4 x i16> @llvm.aarch64.neon.sabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
         %tmp4.1 = zext <4 x i16> %tmp4 to <4 x i32>
         %tmp5 = add <4 x i32> %tmp3, %tmp4.1
@@ -495,9 +561,9 @@ define <4 x i32> @sabal4s(<4 x i16>* %A, <4 x i16>* %B, <4 x i32>* %C) nounwind
 define <2 x i64> @sabal2d(<2 x i32>* %A, <2 x i32>* %B, <2 x i64>* %C) nounwind {
 ;CHECK-LABEL: sabal2d:
 ;CHECK: sabal.2d
-        %tmp1 = load <2 x i32>* %A
-        %tmp2 = load <2 x i32>* %B
-        %tmp3 = load <2 x i64>* %C
+        %tmp1 = load <2 x i32>, <2 x i32>* %A
+        %tmp2 = load <2 x i32>, <2 x i32>* %B
+        %tmp3 = load <2 x i64>, <2 x i64>* %C
         %tmp4 = call <2 x i32> @llvm.aarch64.neon.sabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
         %tmp4.1 = zext <2 x i32> %tmp4 to <2 x i64>
         %tmp4.1.1 = zext <2 x i32> %tmp4 to <2 x i64>
@@ -508,9 +574,9 @@ define <2 x i64> @sabal2d(<2 x i32>* %A, <2 x i32>* %B, <2 x i64>* %C) nounwind
 define <8 x i16> @sabal2_8h(<16 x i8>* %A, <16 x i8>* %B, <8 x i16>* %C) nounwind {
 ;CHECK-LABEL: sabal2_8h:
 ;CHECK: sabal2.8h
-        %load1 = load <16 x i8>* %A
-        %load2 = load <16 x i8>* %B
-        %tmp3 = load <8 x i16>* %C
+        %load1 = load <16 x i8>, <16 x i8>* %A
+        %load2 = load <16 x i8>, <16 x i8>* %B
+        %tmp3 = load <8 x i16>, <8 x i16>* %C
         %tmp1 = shufflevector <16 x i8> %load1, <16 x i8> undef, <8 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
         %tmp2 = shufflevector <16 x i8> %load2, <16 x i8> undef, <8 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
         %tmp4 = call <8 x i8> @llvm.aarch64.neon.sabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
@@ -522,9 +588,9 @@ define <8 x i16> @sabal2_8h(<16 x i8>* %A, <16 x i8>* %B, <8 x i16>* %C) nounwin
 define <4 x i32> @sabal2_4s(<8 x i16>* %A, <8 x i16>* %B, <4 x i32>* %C) nounwind {
 ;CHECK-LABEL: sabal2_4s:
 ;CHECK: sabal2.4s
-        %load1 = load <8 x i16>* %A
-        %load2 = load <8 x i16>* %B
-        %tmp3 = load <4 x i32>* %C
+        %load1 = load <8 x i16>, <8 x i16>* %A
+        %load2 = load <8 x i16>, <8 x i16>* %B
+        %tmp3 = load <4 x i32>, <4 x i32>* %C
         %tmp1 = shufflevector <8 x i16> %load1, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
         %tmp2 = shufflevector <8 x i16> %load2, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
         %tmp4 = call <4 x i16> @llvm.aarch64.neon.sabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
@@ -536,9 +602,9 @@ define <4 x i32> @sabal2_4s(<8 x i16>* %A, <8 x i16>* %B, <4 x i32>* %C) nounwin
 define <2 x i64> @sabal2_2d(<4 x i32>* %A, <4 x i32>* %B, <2 x i64>* %C) nounwind {
 ;CHECK-LABEL: sabal2_2d:
 ;CHECK: sabal2.2d
-        %load1 = load <4 x i32>* %A
-        %load2 = load <4 x i32>* %B
-        %tmp3 = load <2 x i64>* %C
+        %load1 = load <4 x i32>, <4 x i32>* %A
+        %load2 = load <4 x i32>, <4 x i32>* %B
+        %tmp3 = load <2 x i64>, <2 x i64>* %C
         %tmp1 = shufflevector <4 x i32> %load1, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
         %tmp2 = shufflevector <4 x i32> %load2, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
         %tmp4 = call <2 x i32> @llvm.aarch64.neon.sabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
@@ -550,9 +616,9 @@ define <2 x i64> @sabal2_2d(<4 x i32>* %A, <4 x i32>* %B, <2 x i64>* %C) nounwin
 define <8 x i16> @uabal8h(<8 x i8>* %A, <8 x i8>* %B,  <8 x i16>* %C) nounwind {
 ;CHECK-LABEL: uabal8h:
 ;CHECK: uabal.8h
-        %tmp1 = load <8 x i8>* %A
-        %tmp2 = load <8 x i8>* %B
-        %tmp3 = load <8 x i16>* %C
+        %tmp1 = load <8 x i8>, <8 x i8>* %A
+        %tmp2 = load <8 x i8>, <8 x i8>* %B
+        %tmp3 = load <8 x i16>, <8 x i16>* %C
         %tmp4 = call <8 x i8> @llvm.aarch64.neon.uabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
         %tmp4.1 = zext <8 x i8> %tmp4 to <8 x i16>
         %tmp5 = add <8 x i16> %tmp3, %tmp4.1
@@ -562,9 +628,9 @@ define <8 x i16> @uabal8h(<8 x i8>* %A, <8 x i8>* %B,  <8 x i16>* %C) nounwind {
 define <4 x i32> @uabal4s(<4 x i16>* %A, <4 x i16>* %B, <4 x i32>* %C) nounwind {
 ;CHECK-LABEL: uabal4s:
 ;CHECK: uabal.4s
-        %tmp1 = load <4 x i16>* %A
-        %tmp2 = load <4 x i16>* %B
-        %tmp3 = load <4 x i32>* %C
+        %tmp1 = load <4 x i16>, <4 x i16>* %A
+        %tmp2 = load <4 x i16>, <4 x i16>* %B
+        %tmp3 = load <4 x i32>, <4 x i32>* %C
         %tmp4 = call <4 x i16> @llvm.aarch64.neon.uabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
         %tmp4.1 = zext <4 x i16> %tmp4 to <4 x i32>
         %tmp5 = add <4 x i32> %tmp3, %tmp4.1
@@ -574,9 +640,9 @@ define <4 x i32> @uabal4s(<4 x i16>* %A, <4 x i16>* %B, <4 x i32>* %C) nounwind
 define <2 x i64> @uabal2d(<2 x i32>* %A, <2 x i32>* %B, <2 x i64>* %C) nounwind {
 ;CHECK-LABEL: uabal2d:
 ;CHECK: uabal.2d
-        %tmp1 = load <2 x i32>* %A
-        %tmp2 = load <2 x i32>* %B
-        %tmp3 = load <2 x i64>* %C
+        %tmp1 = load <2 x i32>, <2 x i32>* %A
+        %tmp2 = load <2 x i32>, <2 x i32>* %B
+        %tmp3 = load <2 x i64>, <2 x i64>* %C
         %tmp4 = call <2 x i32> @llvm.aarch64.neon.uabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
         %tmp4.1 = zext <2 x i32> %tmp4 to <2 x i64>
         %tmp5 = add <2 x i64> %tmp3, %tmp4.1
@@ -586,9 +652,9 @@ define <2 x i64> @uabal2d(<2 x i32>* %A, <2 x i32>* %B, <2 x i64>* %C) nounwind
 define <8 x i16> @uabal2_8h(<16 x i8>* %A, <16 x i8>* %B, <8 x i16>* %C) nounwind {
 ;CHECK-LABEL: uabal2_8h:
 ;CHECK: uabal2.8h
-        %load1 = load <16 x i8>* %A
-        %load2 = load <16 x i8>* %B
-        %tmp3 = load <8 x i16>* %C
+        %load1 = load <16 x i8>, <16 x i8>* %A
+        %load2 = load <16 x i8>, <16 x i8>* %B
+        %tmp3 = load <8 x i16>, <8 x i16>* %C
         %tmp1 = shufflevector <16 x i8> %load1, <16 x i8> undef, <8 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
         %tmp2 = shufflevector <16 x i8> %load2, <16 x i8> undef, <8 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
         %tmp4 = call <8 x i8> @llvm.aarch64.neon.uabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
@@ -600,9 +666,9 @@ define <8 x i16> @uabal2_8h(<16 x i8>* %A, <16 x i8>* %B, <8 x i16>* %C) nounwin
 define <4 x i32> @uabal2_4s(<8 x i16>* %A, <8 x i16>* %B, <4 x i32>* %C) nounwind {
 ;CHECK-LABEL: uabal2_4s:
 ;CHECK: uabal2.4s
-        %load1 = load <8 x i16>* %A
-        %load2 = load <8 x i16>* %B
-        %tmp3 = load <4 x i32>* %C
+        %load1 = load <8 x i16>, <8 x i16>* %A
+        %load2 = load <8 x i16>, <8 x i16>* %B
+        %tmp3 = load <4 x i32>, <4 x i32>* %C
         %tmp1 = shufflevector <8 x i16> %load1, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
         %tmp2 = shufflevector <8 x i16> %load2, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
         %tmp4 = call <4 x i16> @llvm.aarch64.neon.uabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
@@ -614,9 +680,9 @@ define <4 x i32> @uabal2_4s(<8 x i16>* %A, <8 x i16>* %B, <4 x i32>* %C) nounwin
 define <2 x i64> @uabal2_2d(<4 x i32>* %A, <4 x i32>* %B, <2 x i64>* %C) nounwind {
 ;CHECK-LABEL: uabal2_2d:
 ;CHECK: uabal2.2d
-        %load1 = load <4 x i32>* %A
-        %load2 = load <4 x i32>* %B
-        %tmp3 = load <2 x i64>* %C
+        %load1 = load <4 x i32>, <4 x i32>* %A
+        %load2 = load <4 x i32>, <4 x i32>* %B
+        %tmp3 = load <2 x i64>, <2 x i64>* %C
         %tmp1 = shufflevector <4 x i32> %load1, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
         %tmp2 = shufflevector <4 x i32> %load2, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
         %tmp4 = call <2 x i32> @llvm.aarch64.neon.uabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
@@ -628,10 +694,10 @@ define <2 x i64> @uabal2_2d(<4 x i32>* %A, <4 x i32>* %B, <2 x i64>* %C) nounwin
 define <8 x i8> @saba_8b(<8 x i8>* %A, <8 x i8>* %B, <8 x i8>* %C) nounwind {
 ;CHECK-LABEL: saba_8b:
 ;CHECK: saba.8b
-        %tmp1 = load <8 x i8>* %A
-        %tmp2 = load <8 x i8>* %B
+        %tmp1 = load <8 x i8>, <8 x i8>* %A
+        %tmp2 = load <8 x i8>, <8 x i8>* %B
         %tmp3 = call <8 x i8> @llvm.aarch64.neon.sabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
-        %tmp4 = load <8 x i8>* %C
+        %tmp4 = load <8 x i8>, <8 x i8>* %C
         %tmp5 = add <8 x i8> %tmp3, %tmp4
         ret <8 x i8> %tmp5
 }
@@ -639,10 +705,10 @@ define <8 x i8> @saba_8b(<8 x i8>* %A, <8 x i8>* %B, <8 x i8>* %C) nounwind {
 define <16 x i8> @saba_16b(<16 x i8>* %A, <16 x i8>* %B, <16 x i8>* %C) nounwind {
 ;CHECK-LABEL: saba_16b:
 ;CHECK: saba.16b
-        %tmp1 = load <16 x i8>* %A
-        %tmp2 = load <16 x i8>* %B
+        %tmp1 = load <16 x i8>, <16 x i8>* %A
+        %tmp2 = load <16 x i8>, <16 x i8>* %B
         %tmp3 = call <16 x i8> @llvm.aarch64.neon.sabd.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
-        %tmp4 = load <16 x i8>* %C
+        %tmp4 = load <16 x i8>, <16 x i8>* %C
         %tmp5 = add <16 x i8> %tmp3, %tmp4
         ret <16 x i8> %tmp5
 }
@@ -650,10 +716,10 @@ define <16 x i8> @saba_16b(<16 x i8>* %A, <16 x i8>* %B, <16 x i8>* %C) nounwind
 define <4 x i16> @saba_4h(<4 x i16>* %A, <4 x i16>* %B, <4 x i16>* %C) nounwind {
 ;CHECK-LABEL: saba_4h:
 ;CHECK: saba.4h
-        %tmp1 = load <4 x i16>* %A
-        %tmp2 = load <4 x i16>* %B
+        %tmp1 = load <4 x i16>, <4 x i16>* %A
+        %tmp2 = load <4 x i16>, <4 x i16>* %B
         %tmp3 = call <4 x i16> @llvm.aarch64.neon.sabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
-        %tmp4 = load <4 x i16>* %C
+        %tmp4 = load <4 x i16>, <4 x i16>* %C
         %tmp5 = add <4 x i16> %tmp3, %tmp4
         ret <4 x i16> %tmp5
 }
@@ -661,10 +727,10 @@ define <4 x i16> @saba_4h(<4 x i16>* %A, <4 x i16>* %B, <4 x i16>* %C) nounwind
 define <8 x i16> @saba_8h(<8 x i16>* %A, <8 x i16>* %B, <8 x i16>* %C) nounwind {
 ;CHECK-LABEL: saba_8h:
 ;CHECK: saba.8h
-        %tmp1 = load <8 x i16>* %A
-        %tmp2 = load <8 x i16>* %B
+        %tmp1 = load <8 x i16>, <8 x i16>* %A
+        %tmp2 = load <8 x i16>, <8 x i16>* %B
         %tmp3 = call <8 x i16> @llvm.aarch64.neon.sabd.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
-        %tmp4 = load <8 x i16>* %C
+        %tmp4 = load <8 x i16>, <8 x i16>* %C
         %tmp5 = add <8 x i16> %tmp3, %tmp4
         ret <8 x i16> %tmp5
 }
@@ -672,10 +738,10 @@ define <8 x i16> @saba_8h(<8 x i16>* %A, <8 x i16>* %B, <8 x i16>* %C) nounwind
 define <2 x i32> @saba_2s(<2 x i32>* %A, <2 x i32>* %B, <2 x i32>* %C) nounwind {
 ;CHECK-LABEL: saba_2s:
 ;CHECK: saba.2s
-        %tmp1 = load <2 x i32>* %A
-        %tmp2 = load <2 x i32>* %B
+        %tmp1 = load <2 x i32>, <2 x i32>* %A
+        %tmp2 = load <2 x i32>, <2 x i32>* %B
         %tmp3 = call <2 x i32> @llvm.aarch64.neon.sabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
-        %tmp4 = load <2 x i32>* %C
+        %tmp4 = load <2 x i32>, <2 x i32>* %C
         %tmp5 = add <2 x i32> %tmp3, %tmp4
         ret <2 x i32> %tmp5
 }
@@ -683,10 +749,10 @@ define <2 x i32> @saba_2s(<2 x i32>* %A, <2 x i32>* %B, <2 x i32>* %C) nounwind
 define <4 x i32> @saba_4s(<4 x i32>* %A, <4 x i32>* %B, <4 x i32>* %C) nounwind {
 ;CHECK-LABEL: saba_4s:
 ;CHECK: saba.4s
-        %tmp1 = load <4 x i32>* %A
-        %tmp2 = load <4 x i32>* %B
+        %tmp1 = load <4 x i32>, <4 x i32>* %A
+        %tmp2 = load <4 x i32>, <4 x i32>* %B
         %tmp3 = call <4 x i32> @llvm.aarch64.neon.sabd.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
-        %tmp4 = load <4 x i32>* %C
+        %tmp4 = load <4 x i32>, <4 x i32>* %C
         %tmp5 = add <4 x i32> %tmp3, %tmp4
         ret <4 x i32> %tmp5
 }
@@ -694,10 +760,10 @@ define <4 x i32> @saba_4s(<4 x i32>* %A, <4 x i32>* %B, <4 x i32>* %C) nounwind
 define <8 x i8> @uaba_8b(<8 x i8>* %A, <8 x i8>* %B, <8 x i8>* %C) nounwind {
 ;CHECK-LABEL: uaba_8b:
 ;CHECK: uaba.8b
-        %tmp1 = load <8 x i8>* %A
-        %tmp2 = load <8 x i8>* %B
+        %tmp1 = load <8 x i8>, <8 x i8>* %A
+        %tmp2 = load <8 x i8>, <8 x i8>* %B
         %tmp3 = call <8 x i8> @llvm.aarch64.neon.uabd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
-        %tmp4 = load <8 x i8>* %C
+        %tmp4 = load <8 x i8>, <8 x i8>* %C
         %tmp5 = add <8 x i8> %tmp3, %tmp4
         ret <8 x i8> %tmp5
 }
@@ -705,10 +771,10 @@ define <8 x i8> @uaba_8b(<8 x i8>* %A, <8 x i8>* %B, <8 x i8>* %C) nounwind {
 define <16 x i8> @uaba_16b(<16 x i8>* %A, <16 x i8>* %B, <16 x i8>* %C) nounwind {
 ;CHECK-LABEL: uaba_16b:
 ;CHECK: uaba.16b
-        %tmp1 = load <16 x i8>* %A
-        %tmp2 = load <16 x i8>* %B
+        %tmp1 = load <16 x i8>, <16 x i8>* %A
+        %tmp2 = load <16 x i8>, <16 x i8>* %B
         %tmp3 = call <16 x i8> @llvm.aarch64.neon.uabd.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
-        %tmp4 = load <16 x i8>* %C
+        %tmp4 = load <16 x i8>, <16 x i8>* %C
         %tmp5 = add <16 x i8> %tmp3, %tmp4
         ret <16 x i8> %tmp5
 }
@@ -716,10 +782,10 @@ define <16 x i8> @uaba_16b(<16 x i8>* %A, <16 x i8>* %B, <16 x i8>* %C) nounwind
 define <4 x i16> @uaba_4h(<4 x i16>* %A, <4 x i16>* %B, <4 x i16>* %C) nounwind {
 ;CHECK-LABEL: uaba_4h:
 ;CHECK: uaba.4h
-        %tmp1 = load <4 x i16>* %A
-        %tmp2 = load <4 x i16>* %B
+        %tmp1 = load <4 x i16>, <4 x i16>* %A
+        %tmp2 = load <4 x i16>, <4 x i16>* %B
         %tmp3 = call <4 x i16> @llvm.aarch64.neon.uabd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
-        %tmp4 = load <4 x i16>* %C
+        %tmp4 = load <4 x i16>, <4 x i16>* %C
         %tmp5 = add <4 x i16> %tmp3, %tmp4
         ret <4 x i16> %tmp5
 }
@@ -727,10 +793,10 @@ define <4 x i16> @uaba_4h(<4 x i16>* %A, <4 x i16>* %B, <4 x i16>* %C) nounwind
 define <8 x i16> @uaba_8h(<8 x i16>* %A, <8 x i16>* %B, <8 x i16>* %C) nounwind {
 ;CHECK-LABEL: uaba_8h:
 ;CHECK: uaba.8h
-        %tmp1 = load <8 x i16>* %A
-        %tmp2 = load <8 x i16>* %B
+        %tmp1 = load <8 x i16>, <8 x i16>* %A
+        %tmp2 = load <8 x i16>, <8 x i16>* %B
         %tmp3 = call <8 x i16> @llvm.aarch64.neon.uabd.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
-        %tmp4 = load <8 x i16>* %C
+        %tmp4 = load <8 x i16>, <8 x i16>* %C
         %tmp5 = add <8 x i16> %tmp3, %tmp4
         ret <8 x i16> %tmp5
 }
@@ -738,10 +804,10 @@ define <8 x i16> @uaba_8h(<8 x i16>* %A, <8 x i16>* %B, <8 x i16>* %C) nounwind
 define <2 x i32> @uaba_2s(<2 x i32>* %A, <2 x i32>* %B, <2 x i32>* %C) nounwind {
 ;CHECK-LABEL: uaba_2s:
 ;CHECK: uaba.2s
-        %tmp1 = load <2 x i32>* %A
-        %tmp2 = load <2 x i32>* %B
+        %tmp1 = load <2 x i32>, <2 x i32>* %A
+        %tmp2 = load <2 x i32>, <2 x i32>* %B
         %tmp3 = call <2 x i32> @llvm.aarch64.neon.uabd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
-        %tmp4 = load <2 x i32>* %C
+        %tmp4 = load <2 x i32>, <2 x i32>* %C
         %tmp5 = add <2 x i32> %tmp3, %tmp4
         ret <2 x i32> %tmp5
 }
@@ -749,10 +815,10 @@ define <2 x i32> @uaba_2s(<2 x i32>* %A, <2 x i32>* %B, <2 x i32>* %C) nounwind
 define <4 x i32> @uaba_4s(<4 x i32>* %A, <4 x i32>* %B, <4 x i32>* %C) nounwind {
 ;CHECK-LABEL: uaba_4s:
 ;CHECK: uaba.4s
-        %tmp1 = load <4 x i32>* %A
-        %tmp2 = load <4 x i32>* %B
+        %tmp1 = load <4 x i32>, <4 x i32>* %A
+        %tmp2 = load <4 x i32>, <4 x i32>* %B
         %tmp3 = call <4 x i32> @llvm.aarch64.neon.uabd.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
-        %tmp4 = load <4 x i32>* %C
+        %tmp4 = load <4 x i32>, <4 x i32>* %C
         %tmp5 = add <4 x i32> %tmp3, %tmp4
         ret <4 x i32> %tmp5
 }