[ARM][MVE][Intrinsics] Add MVE VAND/VORR/VORN/VEOR/VBIC intrinsics.

[MarkMurrayARM]

Summary

Add MVE VAND/VORR/VORN/VEOR/VBIC intrinsics.

[MarkMurrayARM]

Rebase.

[MarkMurrayARM]

Rebase.

[MarkMurrayARM]

Rebase.

[MarkMurrayARM]

Rebase and reupload patches.

[simon_tatham]

clang/include/clang/Basic/arm_mve.td

68

I think it's worth adding a comment here explaining why this bitcasting isn't overcomplicating the IR in the integer case. (Because Vector and UVector may be different vector types at the C level – vectors of signed/unsigned ints of the same size – but once they're lowered to IR, they're just vectors of i16 or i32 or whatever, with no sign at all, so the bitcasts will be automatically elided by IRBuilder.)

[MarkMurrayARM]

Respond to review comments.

[simon_tatham]

LGTM, though I spotted a couple of even tinier last-minute nits.

clang/include/clang/Basic/arm_mve.td

68

Typos: "ame" and "igned" should be "same" and "signed".

72

Another long line.

clang/include/clang/Basic/arm_mve.td

foreach n = [ 2, 4 ] in {
   def "vst"#n#"q": Intrinsic<Void, (args Ptr<Scalar>, MultiVector<n>),
                              (CustomCodegen<"VST24"> n:$NumVectors,
                               "Intrinsic::arm_mve_vst"#n#"q":$IRIntr)>;
   def "vld"#n#"q": Intrinsic<MultiVector<n>, (args CPtr<Scalar>),
                              (CustomCodegen<"VLD24"> n:$NumVectors,
                               "Intrinsic::arm_mve_vld"#n#"q":$IRIntr)>;
 }
 
+multiclass bit_op_fp<IRBuilder bitop> {
+def "": Intrinsic<Vector, (args Vector:$a, Vector:$b),
+   (bitcast (bitop (bitcast $a, UVector), (bitcast $b, UVector)), Vector)>;
+}
+
+multiclass bit_op_fp_with_inv<IRBuilder bitop> {
+def "": Intrinsic<Vector, (args Vector:$a, Vector:$b),
+   (bitcast (bitop (bitcast $a, UVector), (not (bitcast $b, UVector))), Vector)>;
+}
 
 let params = T.Int in {
 def vaddq: Intrinsic<Vector, (args Vector:$a, Vector:$b), (add $a, $b)>;
+def vandq: Intrinsic<Vector, (args Vector:$a, Vector:$b), (and $a, $b)>;
+def vbicq: Intrinsic<Vector, (args Vector:$a, Vector:$b), (and $a, (not $b))>;
+def veorq: Intrinsic<Vector, (args Vector:$a, Vector:$b), (xor $a, $b)>;
+def vornq: Intrinsic<Vector, (args Vector:$a, Vector:$b), (or $a, (not $b))>;
+def vorrq: Intrinsic<Vector, (args Vector:$a, Vector:$b), (or $a, $b)>;
 def vsubq: Intrinsic<Vector, (args Vector:$a, Vector:$b), (sub $a, $b)>;
 def vmulq: Intrinsic<Vector, (args Vector:$a, Vector:$b), (mul $a, $b)>;
 }
 
 let params = T.Float in {
 def vaddqf: Intrinsic<Vector, (args Vector:$a, Vector:$b), (fadd $a, $b)>,
             NameOverride<"vaddq">;
+defm vandqf: bit_op_fp<and>, NameOverride<"vandq">;
+defm vbicqf: bit_op_fp_with_inv<and>, NameOverride<"vbicq">;
+defm veorqf: bit_op_fp<xor>, NameOverride<"veorq">;
+defm vornqf: bit_op_fp_with_inv<or>, NameOverride<"vornq">;
+defm vorrqf: bit_op_fp<or>, NameOverride<"vorrq">;
 def vsubqf: Intrinsic<Vector, (args Vector:$a, Vector:$b), (fsub $a, $b)>,
             NameOverride<"vsubq">;
 def vmulqf: Intrinsic<Vector, (args Vector:$a, Vector:$b), (fmul $a, $b)>,
             NameOverride<"vmulq">;
 }
 
+multiclass predicated_bit_op_fp<string int_op> {
+def "": Intrinsic<Vector, (args Vector:$inactive, Vector:$a, Vector:$b, Predicate:$pred),
+    (bitcast (IRInt<int_op, [UVector, Predicate]>
+                    (bitcast $a, UVector),
+                    (bitcast $b, UVector),
+                    $pred,
+                    (bitcast $inactive, UVector)), Vector)>;
+}
+
+// Plain intrinsics
 let params = T.Usual in {
 def vabdq: Intrinsic<Vector, (args Vector:$a, Vector:$b), (IRInt<"vabd", [Vector]> $a, $b)>;
 }
 
+// Predicated intrinsics
 let params = T.Usual in {
 def vabdq_m: Intrinsic<
     Vector, (args Vector:$inactive, Vector:$a, Vector:$b, Predicate:$pred),
     (IRInt<"abd_predicated", [Vector, Predicate]> $a, $b, $pred, $inactive)>;
 def vaddq_m: Intrinsic<
     Vector, (args Vector:$inactive, Vector:$a, Vector:$b, Predicate:$pred),
     (IRInt<"add_predicated", [Vector, Predicate]> $a, $b, $pred, $inactive)>;
 def vsubq_m: Intrinsic<
     Vector, (args Vector:$inactive, Vector:$a, Vector:$b, Predicate:$pred),
     (IRInt<"sub_predicated", [Vector, Predicate]> $a, $b, $pred, $inactive)>;
 def vmulq_m: Intrinsic<
     Vector, (args Vector:$inactive, Vector:$a, Vector:$b, Predicate:$pred),
     (IRInt<"mul_predicated", [Vector, Predicate]> $a, $b, $pred, $inactive)>;
+defm vandq_m: predicated_bit_op_fp<"and_predicated">;
+defm vbicq_m: predicated_bit_op_fp<"bic_predicated">;
+defm veorq_m: predicated_bit_op_fp<"eor_predicated">;
+defm vornq_m: predicated_bit_op_fp<"orn_predicated">;
+defm vorrq_m: predicated_bit_op_fp<"orr_predicated">;
 }
 
 let params = T.Int in {
 def vminvq: Intrinsic<Scalar, (args Scalar:$prev, Vector:$vec),
     (Scalar (IRInt<"minv", [Vector], 1> $prev, $vec))>;
 def vmaxvq: Intrinsic<Scalar, (args Scalar:$prev, Vector:$vec),
     (Scalar (IRInt<"maxv", [Vector], 1> $prev, $vec))>;
 }

clang/include/clang/Basic/arm_mve_defs.td

 }
 class CGHelperFn<string func> : IRBuilderBase {
   // A helper function defined in CGBuiltin.cpp, which takes the IRBuilder as
   // an argument.
   let prefix = func # "(Builder, ";
 }
 def add: IRBuilder<"CreateAdd">;
 def mul: IRBuilder<"CreateMul">;
+def not: IRBuilder<"CreateNot">;
 def or: IRBuilder<"CreateOr">;
 def and: IRBuilder<"CreateAnd">;
+def xor: IRBuilder<"CreateXor">;
 def sub: IRBuilder<"CreateSub">;
 def shl: IRBuilder<"CreateShl">;
 def lshr: IRBuilder<"CreateLShr">;
 def fadd: IRBuilder<"CreateFAdd">;
 def fmul: IRBuilder<"CreateFMul">;
 def fsub: IRBuilder<"CreateFSub">;
 def load: IRBuilder<"CreateLoad"> {
   let special_params = [IRBuilderAddrParam<0>];

clang/test/CodeGen/arm-mve-intrinsics/vandq.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// RUN: %clang_cc1 -triple thumbv8.1m.main-arm-none-eabi -target-feature +mve.fp -mfloat-abi hard -fallow-half-arguments-and-returns -O0 -disable-O0-optnone -S -emit-llvm -o - %s | opt -S -mem2reg | FileCheck %s
+// RUN: %clang_cc1 -triple thumbv8.1m.main-arm-none-eabi -target-feature +mve.fp -mfloat-abi hard -fallow-half-arguments-and-returns -O0 -disable-O0-optnone -DPOLYMORPHIC -S -emit-llvm -o - %s | opt -S -mem2reg | FileCheck %s
+
+#include <arm_mve.h>
+
+// CHECK-LABEL: @test_vandq_u32(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = and <4 x i32> [[A:%.*]], [[B:%.*]]
+// CHECK-NEXT:    ret <4 x i32> [[TMP0]]
+//
+uint32x4_t test_vandq_u32(uint32x4_t a, uint32x4_t b)
+{
+#ifdef POLYMORPHIC
+    return vandq(a, b);
+#else /* POLYMORPHIC */
+    return vandq_u32(a, b);
+#endif /* POLYMORPHIC */
+}
+
+// CHECK-LABEL: @test_vandq_f32(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = bitcast <4 x float> [[A:%.*]] to <4 x i32>
+// CHECK-NEXT:    [[TMP1:%.*]] = bitcast <4 x float> [[B:%.*]] to <4 x i32>
+// CHECK-NEXT:    [[TMP2:%.*]] = and <4 x i32> [[TMP0]], [[TMP1]]
+// CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i32> [[TMP2]] to <4 x float>
+// CHECK-NEXT:    ret <4 x float> [[TMP3]]
+//
+float32x4_t test_vandq_f32(float32x4_t a, float32x4_t b)
+{
+#ifdef POLYMORPHIC
+    return vandq(a, b);
+#else /* POLYMORPHIC */
+    return vandq_f32(a, b);
+#endif /* POLYMORPHIC */
+}
+
+// CHECK-LABEL: @test_vandq_m_s8(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = zext i16 [[P:%.*]] to i32
+// CHECK-NEXT:    [[TMP1:%.*]] = call <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32 [[TMP0]])
+// CHECK-NEXT:    [[TMP2:%.*]] = call <16 x i8> @llvm.arm.mve.and.predicated.v16i8.v16i1(<16 x i8> [[A:%.*]], <16 x i8> [[B:%.*]], <16 x i1> [[TMP1]], <16 x i8> [[INACTIVE:%.*]])
+// CHECK-NEXT:    ret <16 x i8> [[TMP2]]
+//
+int8x16_t test_vandq_m_s8(int8x16_t inactive, int8x16_t a, int8x16_t b, mve_pred16_t p)
+{
+#ifdef POLYMORPHIC
+    return vandq_m(inactive, a, b, p);
+#else /* POLYMORPHIC */
+    return vandq_m_s8(inactive, a, b, p);
+#endif /* POLYMORPHIC */
+}
+
+// CHECK-LABEL: @test_vandq_m_f16(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = bitcast <8 x half> [[A:%.*]] to <8 x i16>
+// CHECK-NEXT:    [[TMP1:%.*]] = bitcast <8 x half> [[B:%.*]] to <8 x i16>
+// CHECK-NEXT:    [[TMP2:%.*]] = zext i16 [[P:%.*]] to i32
+// CHECK-NEXT:    [[TMP3:%.*]] = call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 [[TMP2]])
+// CHECK-NEXT:    [[TMP4:%.*]] = bitcast <8 x half> [[INACTIVE:%.*]] to <8 x i16>
+// CHECK-NEXT:    [[TMP5:%.*]] = call <8 x i16> @llvm.arm.mve.and.predicated.v8i16.v8i1(<8 x i16> [[TMP0]], <8 x i16> [[TMP1]], <8 x i1> [[TMP3]], <8 x i16> [[TMP4]])
+// CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <8 x half>
+// CHECK-NEXT:    ret <8 x half> [[TMP6]]
+//
+float16x8_t test_vandq_m_f16(float16x8_t inactive, float16x8_t a, float16x8_t b, mve_pred16_t p)
+{
+#ifdef POLYMORPHIC
+    return vandq_m(inactive, a, b, p);
+#else /* POLYMORPHIC */
+    return vandq_m_f16(inactive, a, b, p);
+#endif /* POLYMORPHIC */
+}

clang/test/CodeGen/arm-mve-intrinsics/vbicq.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// RUN: %clang_cc1 -triple thumbv8.1m.main-arm-none-eabi -target-feature +mve.fp -mfloat-abi hard -fallow-half-arguments-and-returns -O0 -disable-O0-optnone -S -emit-llvm -o - %s | opt -S -mem2reg | FileCheck %s
+// RUN: %clang_cc1 -triple thumbv8.1m.main-arm-none-eabi -target-feature +mve.fp -mfloat-abi hard -fallow-half-arguments-and-returns -O0 -disable-O0-optnone -DPOLYMORPHIC -S -emit-llvm -o - %s | opt -S -mem2reg | FileCheck %s
+
+#include <arm_mve.h>
+
+// CHECK-LABEL: @test_vbicq_u32(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = xor <4 x i32> [[B:%.*]], <i32 -1, i32 -1, i32 -1, i32 -1>
+// CHECK-NEXT:    [[TMP1:%.*]] = and <4 x i32> [[A:%.*]], [[TMP0]]
+// CHECK-NEXT:    ret <4 x i32> [[TMP1]]
+//
+uint32x4_t test_vbicq_u32(uint32x4_t a, uint32x4_t b)
+{
+#ifdef POLYMORPHIC
+    return vbicq(a, b);
+#else /* POLYMORPHIC */
+    return vbicq_u32(a, b);
+#endif /* POLYMORPHIC */
+}
+
+// CHECK-LABEL: @test_vbicq_f32(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = bitcast <4 x float> [[A:%.*]] to <4 x i32>
+// CHECK-NEXT:    [[TMP1:%.*]] = bitcast <4 x float> [[B:%.*]] to <4 x i32>
+// CHECK-NEXT:    [[TMP2:%.*]] = xor <4 x i32> [[TMP1]], <i32 -1, i32 -1, i32 -1, i32 -1>
+// CHECK-NEXT:    [[TMP3:%.*]] = and <4 x i32> [[TMP0]], [[TMP2]]
+// CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i32> [[TMP3]] to <4 x float>
+// CHECK-NEXT:    ret <4 x float> [[TMP4]]
+//
+float32x4_t test_vbicq_f32(float32x4_t a, float32x4_t b)
+{
+#ifdef POLYMORPHIC
+    return vbicq(a, b);
+#else /* POLYMORPHIC */
+    return vbicq_f32(a, b);
+#endif /* POLYMORPHIC */
+}
+
+// CHECK-LABEL: @test_vbicq_m_s8(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = zext i16 [[P:%.*]] to i32
+// CHECK-NEXT:    [[TMP1:%.*]] = call <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32 [[TMP0]])
+// CHECK-NEXT:    [[TMP2:%.*]] = call <16 x i8> @llvm.arm.mve.bic.predicated.v16i8.v16i1(<16 x i8> [[A:%.*]], <16 x i8> [[B:%.*]], <16 x i1> [[TMP1]], <16 x i8> [[INACTIVE:%.*]])
+// CHECK-NEXT:    ret <16 x i8> [[TMP2]]
+//
+int8x16_t test_vbicq_m_s8(int8x16_t inactive, int8x16_t a, int8x16_t b, mve_pred16_t p)
+{
+#ifdef POLYMORPHIC
+    return vbicq_m(inactive, a, b, p);
+#else /* POLYMORPHIC */
+    return vbicq_m_s8(inactive, a, b, p);
+#endif /* POLYMORPHIC */
+}
+
+// CHECK-LABEL: @test_vbicq_m_f16(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = bitcast <8 x half> [[A:%.*]] to <8 x i16>
+// CHECK-NEXT:    [[TMP1:%.*]] = bitcast <8 x half> [[B:%.*]] to <8 x i16>
+// CHECK-NEXT:    [[TMP2:%.*]] = zext i16 [[P:%.*]] to i32
+// CHECK-NEXT:    [[TMP3:%.*]] = call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 [[TMP2]])
+// CHECK-NEXT:    [[TMP4:%.*]] = bitcast <8 x half> [[INACTIVE:%.*]] to <8 x i16>
+// CHECK-NEXT:    [[TMP5:%.*]] = call <8 x i16> @llvm.arm.mve.bic.predicated.v8i16.v8i1(<8 x i16> [[TMP0]], <8 x i16> [[TMP1]], <8 x i1> [[TMP3]], <8 x i16> [[TMP4]])
+// CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <8 x half>
+// CHECK-NEXT:    ret <8 x half> [[TMP6]]
+//
+float16x8_t test_vbicq_m_f16(float16x8_t inactive, float16x8_t a, float16x8_t b, mve_pred16_t p)
+{
+#ifdef POLYMORPHIC
+    return vbicq_m(inactive, a, b, p);
+#else /* POLYMORPHIC */
+    return vbicq_m_f16(inactive, a, b, p);
+#endif /* POLYMORPHIC */
+}

clang/test/CodeGen/arm-mve-intrinsics/veorq.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// RUN: %clang_cc1 -triple thumbv8.1m.main-arm-none-eabi -target-feature +mve.fp -mfloat-abi hard -fallow-half-arguments-and-returns -O0 -disable-O0-optnone -S -emit-llvm -o - %s | opt -S -mem2reg | FileCheck %s
+// RUN: %clang_cc1 -triple thumbv8.1m.main-arm-none-eabi -target-feature +mve.fp -mfloat-abi hard -fallow-half-arguments-and-returns -O0 -disable-O0-optnone -DPOLYMORPHIC -S -emit-llvm -o - %s | opt -S -mem2reg | FileCheck %s
+
+#include <arm_mve.h>
+
+// CHECK-LABEL: @test_veorq_u32(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = xor <4 x i32> [[A:%.*]], [[B:%.*]]
+// CHECK-NEXT:    ret <4 x i32> [[TMP0]]
+//
+uint32x4_t test_veorq_u32(uint32x4_t a, uint32x4_t b)
+{
+#ifdef POLYMORPHIC
+    return veorq(a, b);
+#else /* POLYMORPHIC */
+    return veorq_u32(a, b);
+#endif /* POLYMORPHIC */
+}
+
+// CHECK-LABEL: @test_veorq_f32(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = bitcast <4 x float> [[A:%.*]] to <4 x i32>
+// CHECK-NEXT:    [[TMP1:%.*]] = bitcast <4 x float> [[B:%.*]] to <4 x i32>
+// CHECK-NEXT:    [[TMP2:%.*]] = xor <4 x i32> [[TMP0]], [[TMP1]]
+// CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i32> [[TMP2]] to <4 x float>
+// CHECK-NEXT:    ret <4 x float> [[TMP3]]
+//
+float32x4_t test_veorq_f32(float32x4_t a, float32x4_t b)
+{
+#ifdef POLYMORPHIC
+    return veorq(a, b);
+#else /* POLYMORPHIC */
+    return veorq_f32(a, b);
+#endif /* POLYMORPHIC */
+}
+
+// CHECK-LABEL: @test_veorq_m_s8(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = zext i16 [[P:%.*]] to i32
+// CHECK-NEXT:    [[TMP1:%.*]] = call <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32 [[TMP0]])
+// CHECK-NEXT:    [[TMP2:%.*]] = call <16 x i8> @llvm.arm.mve.eor.predicated.v16i8.v16i1(<16 x i8> [[A:%.*]], <16 x i8> [[B:%.*]], <16 x i1> [[TMP1]], <16 x i8> [[INACTIVE:%.*]])
+// CHECK-NEXT:    ret <16 x i8> [[TMP2]]
+//
+int8x16_t test_veorq_m_s8(int8x16_t inactive, int8x16_t a, int8x16_t b, mve_pred16_t p)
+{
+#ifdef POLYMORPHIC
+    return veorq_m(inactive, a, b, p);
+#else /* POLYMORPHIC */
+    return veorq_m_s8(inactive, a, b, p);
+#endif /* POLYMORPHIC */
+}
+
+// CHECK-LABEL: @test_veorq_m_f16(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = bitcast <8 x half> [[A:%.*]] to <8 x i16>
+// CHECK-NEXT:    [[TMP1:%.*]] = bitcast <8 x half> [[B:%.*]] to <8 x i16>
+// CHECK-NEXT:    [[TMP2:%.*]] = zext i16 [[P:%.*]] to i32
+// CHECK-NEXT:    [[TMP3:%.*]] = call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 [[TMP2]])
+// CHECK-NEXT:    [[TMP4:%.*]] = bitcast <8 x half> [[INACTIVE:%.*]] to <8 x i16>
+// CHECK-NEXT:    [[TMP5:%.*]] = call <8 x i16> @llvm.arm.mve.eor.predicated.v8i16.v8i1(<8 x i16> [[TMP0]], <8 x i16> [[TMP1]], <8 x i1> [[TMP3]], <8 x i16> [[TMP4]])
+// CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <8 x half>
+// CHECK-NEXT:    ret <8 x half> [[TMP6]]
+//
+float16x8_t test_veorq_m_f16(float16x8_t inactive, float16x8_t a, float16x8_t b, mve_pred16_t p)
+{
+#ifdef POLYMORPHIC
+    return veorq_m(inactive, a, b, p);
+#else /* POLYMORPHIC */
+    return veorq_m_f16(inactive, a, b, p);
+#endif /* POLYMORPHIC */
+}

clang/test/CodeGen/arm-mve-intrinsics/vornq.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// RUN: %clang_cc1 -triple thumbv8.1m.main-arm-none-eabi -target-feature +mve.fp -mfloat-abi hard -fallow-half-arguments-and-returns -O0 -disable-O0-optnone -S -emit-llvm -o - %s | opt -S -mem2reg | FileCheck %s
+// RUN: %clang_cc1 -triple thumbv8.1m.main-arm-none-eabi -target-feature +mve.fp -mfloat-abi hard -fallow-half-arguments-and-returns -O0 -disable-O0-optnone -DPOLYMORPHIC -S -emit-llvm -o - %s | opt -S -mem2reg | FileCheck %s
+
+#include <arm_mve.h>
+
+// CHECK-LABEL: @test_vornq_u32(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = xor <4 x i32> [[B:%.*]], <i32 -1, i32 -1, i32 -1, i32 -1>
+// CHECK-NEXT:    [[TMP1:%.*]] = or <4 x i32> [[A:%.*]], [[TMP0]]
+// CHECK-NEXT:    ret <4 x i32> [[TMP1]]
+//
+uint32x4_t test_vornq_u32(uint32x4_t a, uint32x4_t b)
+{
+#ifdef POLYMORPHIC
+    return vornq(a, b);
+#else /* POLYMORPHIC */
+    return vornq_u32(a, b);
+#endif /* POLYMORPHIC */
+}
+
+// CHECK-LABEL: @test_vornq_f32(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = bitcast <4 x float> [[A:%.*]] to <4 x i32>
+// CHECK-NEXT:    [[TMP1:%.*]] = bitcast <4 x float> [[B:%.*]] to <4 x i32>
+// CHECK-NEXT:    [[TMP2:%.*]] = xor <4 x i32> [[TMP1]], <i32 -1, i32 -1, i32 -1, i32 -1>
+// CHECK-NEXT:    [[TMP3:%.*]] = or <4 x i32> [[TMP0]], [[TMP2]]
+// CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i32> [[TMP3]] to <4 x float>
+// CHECK-NEXT:    ret <4 x float> [[TMP4]]
+//
+float32x4_t test_vornq_f32(float32x4_t a, float32x4_t b)
+{
+#ifdef POLYMORPHIC
+    return vornq(a, b);
+#else /* POLYMORPHIC */
+    return vornq_f32(a, b);
+#endif /* POLYMORPHIC */
+}
+
+// CHECK-LABEL: @test_vornq_m_s8(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = zext i16 [[P:%.*]] to i32
+// CHECK-NEXT:    [[TMP1:%.*]] = call <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32 [[TMP0]])
+// CHECK-NEXT:    [[TMP2:%.*]] = call <16 x i8> @llvm.arm.mve.orn.predicated.v16i8.v16i1(<16 x i8> [[A:%.*]], <16 x i8> [[B:%.*]], <16 x i1> [[TMP1]], <16 x i8> [[INACTIVE:%.*]])
+// CHECK-NEXT:    ret <16 x i8> [[TMP2]]
+//
+int8x16_t test_vornq_m_s8(int8x16_t inactive, int8x16_t a, int8x16_t b, mve_pred16_t p)
+{
+#ifdef POLYMORPHIC
+    return vornq_m(inactive, a, b, p);
+#else /* POLYMORPHIC */
+    return vornq_m_s8(inactive, a, b, p);
+#endif /* POLYMORPHIC */
+}
+
+// CHECK-LABEL: @test_vornq_m_f16(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = bitcast <8 x half> [[A:%.*]] to <8 x i16>
+// CHECK-NEXT:    [[TMP1:%.*]] = bitcast <8 x half> [[B:%.*]] to <8 x i16>
+// CHECK-NEXT:    [[TMP2:%.*]] = zext i16 [[P:%.*]] to i32
+// CHECK-NEXT:    [[TMP3:%.*]] = call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 [[TMP2]])
+// CHECK-NEXT:    [[TMP4:%.*]] = bitcast <8 x half> [[INACTIVE:%.*]] to <8 x i16>
+// CHECK-NEXT:    [[TMP5:%.*]] = call <8 x i16> @llvm.arm.mve.orn.predicated.v8i16.v8i1(<8 x i16> [[TMP0]], <8 x i16> [[TMP1]], <8 x i1> [[TMP3]], <8 x i16> [[TMP4]])
+// CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <8 x half>
+// CHECK-NEXT:    ret <8 x half> [[TMP6]]
+//
+float16x8_t test_vornq_m_f16(float16x8_t inactive, float16x8_t a, float16x8_t b, mve_pred16_t p)
+{
+#ifdef POLYMORPHIC
+    return vornq_m(inactive, a, b, p);
+#else /* POLYMORPHIC */
+    return vornq_m_f16(inactive, a, b, p);
+#endif /* POLYMORPHIC */
+}

clang/test/CodeGen/arm-mve-intrinsics/vorrq.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// RUN: %clang_cc1 -triple thumbv8.1m.main-arm-none-eabi -target-feature +mve.fp -mfloat-abi hard -fallow-half-arguments-and-returns -O0 -disable-O0-optnone -S -emit-llvm -o - %s | opt -S -mem2reg | FileCheck %s
+// RUN: %clang_cc1 -triple thumbv8.1m.main-arm-none-eabi -target-feature +mve.fp -mfloat-abi hard -fallow-half-arguments-and-returns -O0 -disable-O0-optnone -DPOLYMORPHIC -S -emit-llvm -o - %s | opt -S -mem2reg | FileCheck %s
+
+#include <arm_mve.h>
+
+// CHECK-LABEL: @test_vorrq_u32(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = or <4 x i32> [[A:%.*]], [[B:%.*]]
+// CHECK-NEXT:    ret <4 x i32> [[TMP0]]
+//
+uint32x4_t test_vorrq_u32(uint32x4_t a, uint32x4_t b)
+{
+#ifdef POLYMORPHIC
+    return vorrq(a, b);
+#else /* POLYMORPHIC */
+    return vorrq_u32(a, b);
+#endif /* POLYMORPHIC */
+}
+
+// CHECK-LABEL: @test_vorrq_f32(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = bitcast <4 x float> [[A:%.*]] to <4 x i32>
+// CHECK-NEXT:    [[TMP1:%.*]] = bitcast <4 x float> [[B:%.*]] to <4 x i32>
+// CHECK-NEXT:    [[TMP2:%.*]] = or <4 x i32> [[TMP0]], [[TMP1]]
+// CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i32> [[TMP2]] to <4 x float>
+// CHECK-NEXT:    ret <4 x float> [[TMP3]]
+//
+float32x4_t test_vorrq_f32(float32x4_t a, float32x4_t b)
+{
+#ifdef POLYMORPHIC
+    return vorrq(a, b);
+#else /* POLYMORPHIC */
+    return vorrq_f32(a, b);
+#endif /* POLYMORPHIC */
+}
+
+// CHECK-LABEL: @test_vorrq_m_s8(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = zext i16 [[P:%.*]] to i32
+// CHECK-NEXT:    [[TMP1:%.*]] = call <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32 [[TMP0]])
+// CHECK-NEXT:    [[TMP2:%.*]] = call <16 x i8> @llvm.arm.mve.orr.predicated.v16i8.v16i1(<16 x i8> [[A:%.*]], <16 x i8> [[B:%.*]], <16 x i1> [[TMP1]], <16 x i8> [[INACTIVE:%.*]])
+// CHECK-NEXT:    ret <16 x i8> [[TMP2]]
+//
+int8x16_t test_vorrq_m_s8(int8x16_t inactive, int8x16_t a, int8x16_t b, mve_pred16_t p)
+{
+#ifdef POLYMORPHIC
+    return vorrq_m(inactive, a, b, p);
+#else /* POLYMORPHIC */
+    return vorrq_m_s8(inactive, a, b, p);
+#endif /* POLYMORPHIC */
+}
+
+// CHECK-LABEL: @test_vorrq_m_f16(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[TMP0:%.*]] = bitcast <8 x half> [[A:%.*]] to <8 x i16>
+// CHECK-NEXT:    [[TMP1:%.*]] = bitcast <8 x half> [[B:%.*]] to <8 x i16>
+// CHECK-NEXT:    [[TMP2:%.*]] = zext i16 [[P:%.*]] to i32
+// CHECK-NEXT:    [[TMP3:%.*]] = call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 [[TMP2]])
+// CHECK-NEXT:    [[TMP4:%.*]] = bitcast <8 x half> [[INACTIVE:%.*]] to <8 x i16>
+// CHECK-NEXT:    [[TMP5:%.*]] = call <8 x i16> @llvm.arm.mve.orr.predicated.v8i16.v8i1(<8 x i16> [[TMP0]], <8 x i16> [[TMP1]], <8 x i1> [[TMP3]], <8 x i16> [[TMP4]])
+// CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <8 x half>
+// CHECK-NEXT:    ret <8 x half> [[TMP6]]
+//
+float16x8_t test_vorrq_m_f16(float16x8_t inactive, float16x8_t a, float16x8_t b, mve_pred16_t p)
+{
+#ifdef POLYMORPHIC
+    return vorrq_m(inactive, a, b, p);
+#else /* POLYMORPHIC */
+    return vorrq_m_f16(inactive, a, b, p);
+#endif /* POLYMORPHIC */
+}

llvm/include/llvm/IR/IntrinsicsARM.td

 }
 
 def int_arm_mve_abd_predicated: Intrinsic<[llvm_anyvector_ty],
    [LLVMMatchType<0>, LLVMMatchType<0>, llvm_anyvector_ty, LLVMMatchType<0>],
    [IntrNoMem]>;
 def int_arm_mve_add_predicated: Intrinsic<[llvm_anyvector_ty],
    [LLVMMatchType<0>, LLVMMatchType<0>, llvm_anyvector_ty, LLVMMatchType<0>],
    [IntrNoMem]>;
+def int_arm_mve_and_predicated: Intrinsic<[llvm_anyvector_ty],
+   [LLVMMatchType<0>, LLVMMatchType<0>, llvm_anyvector_ty, LLVMMatchType<0>],
+   [IntrNoMem]>;
+def int_arm_mve_bic_predicated: Intrinsic<[llvm_anyvector_ty],
+   [LLVMMatchType<0>, LLVMMatchType<0>, llvm_anyvector_ty, LLVMMatchType<0>],
+   [IntrNoMem]>;
+def int_arm_mve_eor_predicated: Intrinsic<[llvm_anyvector_ty],
+   [LLVMMatchType<0>, LLVMMatchType<0>, llvm_anyvector_ty, LLVMMatchType<0>],
+   [IntrNoMem]>;
+def int_arm_mve_orn_predicated: Intrinsic<[llvm_anyvector_ty],
+   [LLVMMatchType<0>, LLVMMatchType<0>, llvm_anyvector_ty, LLVMMatchType<0>],
+   [IntrNoMem]>;
+def int_arm_mve_orr_predicated: Intrinsic<[llvm_anyvector_ty],
+   [LLVMMatchType<0>, LLVMMatchType<0>, llvm_anyvector_ty, LLVMMatchType<0>],
+   [IntrNoMem]>;
 def int_arm_mve_sub_predicated: Intrinsic<[llvm_anyvector_ty],
    [LLVMMatchType<0>, LLVMMatchType<0>, llvm_anyvector_ty, LLVMMatchType<0>],
    [IntrNoMem]>;
 def int_arm_mve_mul_predicated: Intrinsic<[llvm_anyvector_ty],
    [LLVMMatchType<0>, LLVMMatchType<0>, llvm_anyvector_ty, LLVMMatchType<0>],
    [IntrNoMem]>;
 
 defm int_arm_mve_minv: IntrinsicSignSuffix<[llvm_i32_ty],

llvm/lib/Target/ARM/ARMInstrMVE.td

foreach s=["s8", "s16", "s32", "u8", "u16", "u32", "i8", "i16", "i32", "f16", "f32"] in {
   def : MVEInstAlias<"vorn${vp}." # s # "\t$QdSrc, $QnSrc, $QmSrc",
         (MVE_VORN MQPR:$QdSrc, MQPR:$QnSrc, MQPR:$QmSrc, vpred_r:$vp)>;
   def : MVEInstAlias<"vorr${vp}." # s # "\t$QdSrc, $QnSrc, $QmSrc",
         (MVE_VORR MQPR:$QdSrc, MQPR:$QnSrc, MQPR:$QmSrc, vpred_r:$vp)>;
   def : MVEInstAlias<"vand${vp}." # s # "\t$QdSrc, $QnSrc, $QmSrc",
         (MVE_VAND MQPR:$QdSrc, MQPR:$QnSrc, MQPR:$QmSrc, vpred_r:$vp)>;
 }
 
+multiclass MVE_bit_op<MVEVectorVTInfo VTI, SDNode unpred_op, Intrinsic pred_int, MVE_bit_ops instruction> {
   let Predicates = [HasMVEInt] in {
-  def : Pat<(v16i8 (and (v16i8 MQPR:$val1), (v16i8 MQPR:$val2))),
-            (v16i8 (MVE_VAND (v16i8 MQPR:$val1), (v16i8 MQPR:$val2)))>;
-  def : Pat<(v8i16 (and (v8i16 MQPR:$val1), (v8i16 MQPR:$val2))),
-            (v8i16 (MVE_VAND (v8i16 MQPR:$val1), (v8i16 MQPR:$val2)))>;
-  def : Pat<(v4i32 (and (v4i32 MQPR:$val1), (v4i32 MQPR:$val2))),
-            (v4i32 (MVE_VAND (v4i32 MQPR:$val1), (v4i32 MQPR:$val2)))>;
-  def : Pat<(v2i64 (and (v2i64 MQPR:$val1), (v2i64 MQPR:$val2))),
-            (v2i64 (MVE_VAND (v2i64 MQPR:$val1), (v2i64 MQPR:$val2)))>;
-
-  def : Pat<(v16i8 (or (v16i8 MQPR:$val1), (v16i8 MQPR:$val2))),
-            (v16i8 (MVE_VORR (v16i8 MQPR:$val1), (v16i8 MQPR:$val2)))>;
-  def : Pat<(v8i16 (or (v8i16 MQPR:$val1), (v8i16 MQPR:$val2))),
-            (v8i16 (MVE_VORR (v8i16 MQPR:$val1), (v8i16 MQPR:$val2)))>;
-  def : Pat<(v4i32 (or (v4i32 MQPR:$val1), (v4i32 MQPR:$val2))),
-            (v4i32 (MVE_VORR (v4i32 MQPR:$val1), (v4i32 MQPR:$val2)))>;
-  def : Pat<(v2i64 (or (v2i64 MQPR:$val1), (v2i64 MQPR:$val2))),
-            (v2i64 (MVE_VORR (v2i64 MQPR:$val1), (v2i64 MQPR:$val2)))>;
-
-  def : Pat<(v16i8 (xor (v16i8 MQPR:$val1), (v16i8 MQPR:$val2))),
-            (v16i8 (MVE_VEOR (v16i8 MQPR:$val1), (v16i8 MQPR:$val2)))>;
-  def : Pat<(v8i16 (xor (v8i16 MQPR:$val1), (v8i16 MQPR:$val2))),
-            (v8i16 (MVE_VEOR (v8i16 MQPR:$val1), (v8i16 MQPR:$val2)))>;
-  def : Pat<(v4i32 (xor (v4i32 MQPR:$val1), (v4i32 MQPR:$val2))),
-            (v4i32 (MVE_VEOR (v4i32 MQPR:$val1), (v4i32 MQPR:$val2)))>;
-  def : Pat<(v2i64 (xor (v2i64 MQPR:$val1), (v2i64 MQPR:$val2))),
-            (v2i64 (MVE_VEOR (v2i64 MQPR:$val1), (v2i64 MQPR:$val2)))>;
-
-  def : Pat<(v16i8 (and (v16i8 MQPR:$val1), (vnotq MQPR:$val2))),
-            (v16i8 (MVE_VBIC (v16i8 MQPR:$val1), (v16i8 MQPR:$val2)))>;
-  def : Pat<(v8i16 (and (v8i16 MQPR:$val1), (vnotq MQPR:$val2))),
-            (v8i16 (MVE_VBIC (v8i16 MQPR:$val1), (v8i16 MQPR:$val2)))>;
-  def : Pat<(v4i32 (and (v4i32 MQPR:$val1), (vnotq MQPR:$val2))),
-            (v4i32 (MVE_VBIC (v4i32 MQPR:$val1), (v4i32 MQPR:$val2)))>;
-  def : Pat<(v2i64 (and (v2i64 MQPR:$val1), (vnotq MQPR:$val2))),
-            (v2i64 (MVE_VBIC (v2i64 MQPR:$val1), (v2i64 MQPR:$val2)))>;
-
-  def : Pat<(v16i8 (or (v16i8 MQPR:$val1), (vnotq MQPR:$val2))),
-            (v16i8 (MVE_VORN (v16i8 MQPR:$val1), (v16i8 MQPR:$val2)))>;
-  def : Pat<(v8i16 (or (v8i16 MQPR:$val1), (vnotq MQPR:$val2))),
-            (v8i16 (MVE_VORN (v8i16 MQPR:$val1), (v8i16 MQPR:$val2)))>;
-  def : Pat<(v4i32 (or (v4i32 MQPR:$val1), (vnotq MQPR:$val2))),
-            (v4i32 (MVE_VORN (v4i32 MQPR:$val1), (v4i32 MQPR:$val2)))>;
-  def : Pat<(v2i64 (or (v2i64 MQPR:$val1), (vnotq MQPR:$val2))),
-            (v2i64 (MVE_VORN (v2i64 MQPR:$val1), (v2i64 MQPR:$val2)))>;
+    // Unpredicated operation
+    def : Pat<(VTI.Vec (unpred_op (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn))),
+              (VTI.Vec (instruction (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn)))>;
+    // Predicated operation
+    def : Pat<(VTI.Vec (pred_int (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn),
+                            (VTI.Pred VCCR:$mask), (VTI.Vec MQPR:$inactive))),
+              (VTI.Vec (instruction
+                            (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn),
+                            (i32 1), (VTI.Pred VCCR:$mask),
+                            (VTI.Vec MQPR:$inactive)))>;
+  }
+}
+
+defm : MVE_bit_op<MVE_v16i8, and, int_arm_mve_and_predicated, MVE_VAND>;
+defm : MVE_bit_op<MVE_v8i16, and, int_arm_mve_and_predicated, MVE_VAND>;
+defm : MVE_bit_op<MVE_v4i32, and, int_arm_mve_and_predicated, MVE_VAND>;
+defm : MVE_bit_op<MVE_v2i64, and, int_arm_mve_and_predicated, MVE_VAND>;
+
+defm : MVE_bit_op<MVE_v16i8, or, int_arm_mve_orr_predicated, MVE_VORR>;
+defm : MVE_bit_op<MVE_v8i16, or, int_arm_mve_orr_predicated, MVE_VORR>;
+defm : MVE_bit_op<MVE_v4i32, or, int_arm_mve_orr_predicated, MVE_VORR>;
+defm : MVE_bit_op<MVE_v2i64, or, int_arm_mve_orr_predicated, MVE_VORR>;
+
+defm : MVE_bit_op<MVE_v16i8, xor, int_arm_mve_eor_predicated, MVE_VEOR>;
+defm : MVE_bit_op<MVE_v8i16, xor, int_arm_mve_eor_predicated, MVE_VEOR>;
+defm : MVE_bit_op<MVE_v4i32, xor, int_arm_mve_eor_predicated, MVE_VEOR>;
+defm : MVE_bit_op<MVE_v2i64, xor, int_arm_mve_eor_predicated, MVE_VEOR>;
+
+multiclass MVE_bit_op_with_inv<MVEVectorVTInfo VTI, SDNode unpred_op, Intrinsic pred_int, MVE_bit_ops instruction> {
+  let Predicates = [HasMVEInt] in {
+    // Unpredicated operation
+    def : Pat<(VTI.Vec (unpred_op (VTI.Vec MQPR:$Qm), (vnotq (VTI.Vec MQPR:$Qn)))),
+              (VTI.Vec (instruction (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn)))>;
+    // Predicated operation
+    def : Pat<(VTI.Vec (pred_int (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn),
+                            (VTI.Pred VCCR:$mask), (VTI.Vec MQPR:$inactive))),
+              (VTI.Vec (instruction
+                            (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn),
+                            (i32 1), (VTI.Pred VCCR:$mask),
+                            (VTI.Vec MQPR:$inactive)))>;
   }
+}
+
+defm : MVE_bit_op_with_inv<MVE_v16i8, and, int_arm_mve_bic_predicated, MVE_VBIC>;
+defm : MVE_bit_op_with_inv<MVE_v8i16, and, int_arm_mve_bic_predicated, MVE_VBIC>;
+defm : MVE_bit_op_with_inv<MVE_v4i32, and, int_arm_mve_bic_predicated, MVE_VBIC>;
+defm : MVE_bit_op_with_inv<MVE_v2i64, and, int_arm_mve_bic_predicated, MVE_VBIC>;
+
+defm : MVE_bit_op_with_inv<MVE_v16i8, or, int_arm_mve_orn_predicated, MVE_VORN>;
+defm : MVE_bit_op_with_inv<MVE_v8i16, or, int_arm_mve_orn_predicated, MVE_VORN>;
+defm : MVE_bit_op_with_inv<MVE_v4i32, or, int_arm_mve_orn_predicated, MVE_VORN>;
+defm : MVE_bit_op_with_inv<MVE_v2i64, or, int_arm_mve_orn_predicated, MVE_VORN>;
 
 class MVE_bit_cmode<string iname, string suffix, bits<4> cmode, dag inOps>
   : MVE_p<(outs MQPR:$Qd), inOps, NoItinerary,
           iname, suffix, "$Qd, $imm", vpred_n, "$Qd = $Qd_src"> {
   bits<8> imm;
   bits<4> Qd;
 
   let Inst{28} = imm{7};

llvm/test/CodeGen/Thumb2/mve-intrinsics/vandq.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=thumbv8.1m.main -mattr=+mve.fp -verify-machineinstrs -o - %s | FileCheck %s
+
+define arm_aapcs_vfpcc <16 x i8> @test_vandq_u8(<16 x i8> %a, <16 x i8> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vandq_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vand q0, q1, q0
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = and <16 x i8> %b, %a
+  ret <16 x i8> %0
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vandq_u32(<4 x i32> %a, <4 x i32> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vandq_u32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vand q0, q1, q0
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = and <4 x i32> %b, %a
+  ret <4 x i32> %0
+}
+
+define arm_aapcs_vfpcc <8 x i16> @test_vandq_s16(<8 x i16> %a, <8 x i16> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vandq_s16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vand q0, q1, q0
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = and <8 x i16> %b, %a
+  ret <8 x i16> %0
+}
+
+define arm_aapcs_vfpcc <4 x float> @test_vandq_f32(<4 x float> %a, <4 x float> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vandq_f32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vand q0, q1, q0
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = bitcast <4 x float> %a to <4 x i32>
+  %1 = bitcast <4 x float> %b to <4 x i32>
+  %2 = and <4 x i32> %1, %0
+  %3 = bitcast <4 x i32> %2 to <4 x float>
+  ret <4 x float> %3
+}
+
+define arm_aapcs_vfpcc <16 x i8> @test_vandq_m_s8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vandq_m_s8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vandt q0, q1, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32 %0)
+  %2 = tail call <16 x i8> @llvm.arm.mve.and.predicated.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, <16 x i1> %1, <16 x i8> %inactive)
+  ret <16 x i8> %2
+}
+
+declare <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32) #2
+
+declare <16 x i8> @llvm.arm.mve.and.predicated.v16i8.v16i1(<16 x i8>, <16 x i8>, <16 x i1>, <16 x i8>) #2
+
+define arm_aapcs_vfpcc <8 x i16> @test_vandq_m_u16(<8 x i16> %inactive, <8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vandq_m_u16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vandt q0, q1, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %0)
+  %2 = tail call <8 x i16> @llvm.arm.mve.and.predicated.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, <8 x i1> %1, <8 x i16> %inactive)
+  ret <8 x i16> %2
+}
+
+declare <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32) #2
+
+declare <8 x i16> @llvm.arm.mve.and.predicated.v8i16.v8i1(<8 x i16>, <8 x i16>, <8 x i1>, <8 x i16>) #2
+
+; Function Attrs: nounwind readnone
+define arm_aapcs_vfpcc <8 x half> @test_vandq_m_f32(<4 x float> %inactive, <4 x float> %a, <4 x float> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vandq_m_f32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vandt q0, q1, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = bitcast <4 x float> %a to <4 x i32>
+  %1 = bitcast <4 x float> %b to <4 x i32>
+  %2 = zext i16 %p to i32
+  %3 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %2)
+  %4 = bitcast <4 x float> %inactive to <4 x i32>
+  %5 = tail call <4 x i32> @llvm.arm.mve.and.predicated.v4i32.v4i1(<4 x i32> %0, <4 x i32> %1, <4 x i1> %3, <4 x i32> %4)
+  %6 = bitcast <4 x i32> %5 to <8 x half>
+  ret <8 x half> %6
+}
+
+declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32) #2
+
+declare <4 x i32> @llvm.arm.mve.and.predicated.v4i32.v4i1(<4 x i32>, <4 x i32>, <4 x i1>, <4 x i32>) #2

llvm/test/CodeGen/Thumb2/mve-intrinsics/vbicq.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=thumbv8.1m.main -mattr=+mve.fp -verify-machineinstrs -o - %s | FileCheck %s
+
+define arm_aapcs_vfpcc <16 x i8> @test_vbicq_u8(<16 x i8> %a, <16 x i8> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vbicq_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vbic q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = xor <16 x i8> %b, <i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1>
+  %1 = and <16 x i8> %0, %a
+  ret <16 x i8> %1
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vbicq_u32(<4 x i32> %a, <4 x i32> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vbicq_u32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vbic q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = xor <4 x i32> %b, <i32 -1, i32 -1, i32 -1, i32 -1>
+  %1 = and <4 x i32> %0, %a
+  ret <4 x i32> %1
+}
+
+define arm_aapcs_vfpcc <8 x i16> @test_vbicq_s16(<8 x i16> %a, <8 x i16> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vbicq_s16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vbic q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = xor <8 x i16> %b, <i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1>
+  %1 = and <8 x i16> %0, %a
+  ret <8 x i16> %1
+}
+
+define arm_aapcs_vfpcc <4 x float> @test_vbicq_f32(<4 x float> %a, <4 x float> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vbicq_f32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vbic q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = bitcast <4 x float> %a to <4 x i32>
+  %1 = bitcast <4 x float> %b to <4 x i32>
+  %2 = xor <4 x i32> %1, <i32 -1, i32 -1, i32 -1, i32 -1>
+  %3 = and <4 x i32> %2, %0
+  %4 = bitcast <4 x i32> %3 to <4 x float>
+  ret <4 x float> %4
+}
+
+define arm_aapcs_vfpcc <16 x i8> @test_vbicq_m_s8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vbicq_m_s8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vbict q0, q1, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32 %0)
+  %2 = tail call <16 x i8> @llvm.arm.mve.bic.predicated.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, <16 x i1> %1, <16 x i8> %inactive)
+  ret <16 x i8> %2
+}
+
+declare <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32) #2
+
+declare <16 x i8> @llvm.arm.mve.bic.predicated.v16i8.v16i1(<16 x i8>, <16 x i8>, <16 x i1>, <16 x i8>) #2
+
+define arm_aapcs_vfpcc <8 x i16> @test_vbicq_m_u16(<8 x i16> %inactive, <8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vbicq_m_u16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vbict q0, q1, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %0)
+  %2 = tail call <8 x i16> @llvm.arm.mve.bic.predicated.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, <8 x i1> %1, <8 x i16> %inactive)
+  ret <8 x i16> %2
+}
+
+declare <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32) #2
+
+declare <8 x i16> @llvm.arm.mve.bic.predicated.v8i16.v8i1(<8 x i16>, <8 x i16>, <8 x i1>, <8 x i16>) #2
+
+; Function Attrs: nounwind readnone
+define arm_aapcs_vfpcc <8 x half> @test_vbicq_m_f32(<4 x float> %inactive, <4 x float> %a, <4 x float> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vbicq_m_f32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vbict q0, q1, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = bitcast <4 x float> %a to <4 x i32>
+  %1 = bitcast <4 x float> %b to <4 x i32>
+  %2 = zext i16 %p to i32
+  %3 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %2)
+  %4 = bitcast <4 x float> %inactive to <4 x i32>
+  %5 = tail call <4 x i32> @llvm.arm.mve.bic.predicated.v4i32.v4i1(<4 x i32> %0, <4 x i32> %1, <4 x i1> %3, <4 x i32> %4)
+  %6 = bitcast <4 x i32> %5 to <8 x half>
+  ret <8 x half> %6
+}
+
+declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32) #2
+
+declare <4 x i32> @llvm.arm.mve.bic.predicated.v4i32.v4i1(<4 x i32>, <4 x i32>, <4 x i1>, <4 x i32>) #2

llvm/test/CodeGen/Thumb2/mve-intrinsics/veorq.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=thumbv8.1m.main -mattr=+mve.fp -verify-machineinstrs -o - %s | FileCheck %s
+
+define arm_aapcs_vfpcc <16 x i8> @test_veorq_u8(<16 x i8> %a, <16 x i8> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_veorq_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    veor q0, q1, q0
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = xor <16 x i8> %b, %a
+  ret <16 x i8> %0
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_veorq_u32(<4 x i32> %a, <4 x i32> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_veorq_u32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    veor q0, q1, q0
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = xor <4 x i32> %b, %a
+  ret <4 x i32> %0
+}
+
+define arm_aapcs_vfpcc <8 x i16> @test_veorq_s16(<8 x i16> %a, <8 x i16> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_veorq_s16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    veor q0, q1, q0
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = xor <8 x i16> %b, %a
+  ret <8 x i16> %0
+}
+
+define arm_aapcs_vfpcc <4 x float> @test_veorq_f32(<4 x float> %a, <4 x float> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_veorq_f32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    veor q0, q1, q0
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = bitcast <4 x float> %a to <4 x i32>
+  %1 = bitcast <4 x float> %b to <4 x i32>
+  %2 = xor <4 x i32> %1, %0
+  %3 = bitcast <4 x i32> %2 to <4 x float>
+  ret <4 x float> %3
+}
+
+define arm_aapcs_vfpcc <16 x i8> @test_veorq_m_s8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_veorq_m_s8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    veort q0, q1, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32 %0)
+  %2 = tail call <16 x i8> @llvm.arm.mve.eor.predicated.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, <16 x i1> %1, <16 x i8> %inactive)
+  ret <16 x i8> %2
+}
+
+declare <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32) #2
+
+declare <16 x i8> @llvm.arm.mve.eor.predicated.v16i8.v16i1(<16 x i8>, <16 x i8>, <16 x i1>, <16 x i8>) #2
+
+define arm_aapcs_vfpcc <8 x i16> @test_veorq_m_u16(<8 x i16> %inactive, <8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_veorq_m_u16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    veort q0, q1, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %0)
+  %2 = tail call <8 x i16> @llvm.arm.mve.eor.predicated.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, <8 x i1> %1, <8 x i16> %inactive)
+  ret <8 x i16> %2
+}
+
+declare <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32) #2
+
+declare <8 x i16> @llvm.arm.mve.eor.predicated.v8i16.v8i1(<8 x i16>, <8 x i16>, <8 x i1>, <8 x i16>) #2
+
+; Function Attrs: nounwind readnone
+define arm_aapcs_vfpcc <8 x half> @test_veorq_m_f32(<4 x float> %inactive, <4 x float> %a, <4 x float> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_veorq_m_f32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    veort q0, q1, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = bitcast <4 x float> %a to <4 x i32>
+  %1 = bitcast <4 x float> %b to <4 x i32>
+  %2 = zext i16 %p to i32
+  %3 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %2)
+  %4 = bitcast <4 x float> %inactive to <4 x i32>
+  %5 = tail call <4 x i32> @llvm.arm.mve.eor.predicated.v4i32.v4i1(<4 x i32> %0, <4 x i32> %1, <4 x i1> %3, <4 x i32> %4)
+  %6 = bitcast <4 x i32> %5 to <8 x half>
+  ret <8 x half> %6
+}
+
+declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32) #2
+
+declare <4 x i32> @llvm.arm.mve.eor.predicated.v4i32.v4i1(<4 x i32>, <4 x i32>, <4 x i1>, <4 x i32>) #2

llvm/test/CodeGen/Thumb2/mve-intrinsics/vornq.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=thumbv8.1m.main -mattr=+mve.fp -verify-machineinstrs -o - %s | FileCheck %s
+
+define arm_aapcs_vfpcc <16 x i8> @test_vornq_u8(<16 x i8> %a, <16 x i8> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vornq_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vorn q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = xor <16 x i8> %b, <i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1>
+  %1 = or <16 x i8> %0, %a
+  ret <16 x i8> %1
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vornq_u32(<4 x i32> %a, <4 x i32> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vornq_u32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vorn q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = xor <4 x i32> %b, <i32 -1, i32 -1, i32 -1, i32 -1>
+  %1 = or <4 x i32> %0, %a
+  ret <4 x i32> %1
+}
+
+define arm_aapcs_vfpcc <8 x i16> @test_vornq_s16(<8 x i16> %a, <8 x i16> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vornq_s16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vorn q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = xor <8 x i16> %b, <i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1>
+  %1 = or <8 x i16> %0, %a
+  ret <8 x i16> %1
+}
+
+define arm_aapcs_vfpcc <4 x float> @test_vornq_f32(<4 x float> %a, <4 x float> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vornq_f32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vorn q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = bitcast <4 x float> %a to <4 x i32>
+  %1 = bitcast <4 x float> %b to <4 x i32>
+  %2 = xor <4 x i32> %1, <i32 -1, i32 -1, i32 -1, i32 -1>
+  %3 = or <4 x i32> %2, %0
+  %4 = bitcast <4 x i32> %3 to <4 x float>
+  ret <4 x float> %4
+}
+
+define arm_aapcs_vfpcc <16 x i8> @test_vornq_m_s8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vornq_m_s8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vornt q0, q1, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32 %0)
+  %2 = tail call <16 x i8> @llvm.arm.mve.orn.predicated.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, <16 x i1> %1, <16 x i8> %inactive)
+  ret <16 x i8> %2
+}
+
+declare <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32) #2
+
+declare <16 x i8> @llvm.arm.mve.orn.predicated.v16i8.v16i1(<16 x i8>, <16 x i8>, <16 x i1>, <16 x i8>) #2
+
+define arm_aapcs_vfpcc <8 x i16> @test_vornq_m_u16(<8 x i16> %inactive, <8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vornq_m_u16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vornt q0, q1, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %0)
+  %2 = tail call <8 x i16> @llvm.arm.mve.orn.predicated.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, <8 x i1> %1, <8 x i16> %inactive)
+  ret <8 x i16> %2
+}
+
+declare <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32) #2
+
+declare <8 x i16> @llvm.arm.mve.orn.predicated.v8i16.v8i1(<8 x i16>, <8 x i16>, <8 x i1>, <8 x i16>) #2
+
+; Function Attrs: nounwind readnone
+define arm_aapcs_vfpcc <8 x half> @test_vornq_m_f32(<4 x float> %inactive, <4 x float> %a, <4 x float> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vornq_m_f32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vornt q0, q1, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = bitcast <4 x float> %a to <4 x i32>
+  %1 = bitcast <4 x float> %b to <4 x i32>
+  %2 = zext i16 %p to i32
+  %3 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %2)
+  %4 = bitcast <4 x float> %inactive to <4 x i32>
+  %5 = tail call <4 x i32> @llvm.arm.mve.orn.predicated.v4i32.v4i1(<4 x i32> %0, <4 x i32> %1, <4 x i1> %3, <4 x i32> %4)
+  %6 = bitcast <4 x i32> %5 to <8 x half>
+  ret <8 x half> %6
+}
+
+declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32) #2
+
+declare <4 x i32> @llvm.arm.mve.orn.predicated.v4i32.v4i1(<4 x i32>, <4 x i32>, <4 x i1>, <4 x i32>) #2

llvm/test/CodeGen/Thumb2/mve-intrinsics/vorrq.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=thumbv8.1m.main -mattr=+mve.fp -verify-machineinstrs -o - %s | FileCheck %s
+
+define arm_aapcs_vfpcc <16 x i8> @test_vorrq_u8(<16 x i8> %a, <16 x i8> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vorrq_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vorr q0, q1, q0
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = or <16 x i8> %b, %a
+  ret <16 x i8> %0
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vorrq_u32(<4 x i32> %a, <4 x i32> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vorrq_u32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vorr q0, q1, q0
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = or <4 x i32> %b, %a
+  ret <4 x i32> %0
+}
+
+define arm_aapcs_vfpcc <8 x i16> @test_vorrq_s16(<8 x i16> %a, <8 x i16> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vorrq_s16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vorr q0, q1, q0
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = or <8 x i16> %b, %a
+  ret <8 x i16> %0
+}
+
+define arm_aapcs_vfpcc <4 x float> @test_vorrq_f32(<4 x float> %a, <4 x float> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vorrq_f32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vorr q0, q1, q0
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = bitcast <4 x float> %a to <4 x i32>
+  %1 = bitcast <4 x float> %b to <4 x i32>
+  %2 = or <4 x i32> %1, %0
+  %3 = bitcast <4 x i32> %2 to <4 x float>
+  ret <4 x float> %3
+}
+
+define arm_aapcs_vfpcc <16 x i8> @test_vorrq_m_s8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vorrq_m_s8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vorrt q0, q1, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32 %0)
+  %2 = tail call <16 x i8> @llvm.arm.mve.orr.predicated.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, <16 x i1> %1, <16 x i8> %inactive)
+  ret <16 x i8> %2
+}
+
+declare <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32) #2
+
+declare <16 x i8> @llvm.arm.mve.orr.predicated.v16i8.v16i1(<16 x i8>, <16 x i8>, <16 x i1>, <16 x i8>) #2
+
+define arm_aapcs_vfpcc <8 x i16> @test_vorrq_m_u16(<8 x i16> %inactive, <8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vorrq_m_u16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vorrt q0, q1, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %0)
+  %2 = tail call <8 x i16> @llvm.arm.mve.orr.predicated.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, <8 x i1> %1, <8 x i16> %inactive)
+  ret <8 x i16> %2
+}
+
+declare <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32) #2
+
+declare <8 x i16> @llvm.arm.mve.orr.predicated.v8i16.v8i1(<8 x i16>, <8 x i16>, <8 x i1>, <8 x i16>) #2
+
+; Function Attrs: nounwind readnone
+define arm_aapcs_vfpcc <8 x half> @test_vorrq_m_f32(<4 x float> %inactive, <4 x float> %a, <4 x float> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vorrq_m_f32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vorrt q0, q1, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = bitcast <4 x float> %a to <4 x i32>
+  %1 = bitcast <4 x float> %b to <4 x i32>
+  %2 = zext i16 %p to i32
+  %3 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %2)
+  %4 = bitcast <4 x float> %inactive to <4 x i32>
+  %5 = tail call <4 x i32> @llvm.arm.mve.orr.predicated.v4i32.v4i1(<4 x i32> %0, <4 x i32> %1, <4 x i1> %3, <4 x i32> %4)
+  %6 = bitcast <4 x i32> %5 to <8 x half>
+  ret <8 x half> %6
+}
+
+declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32) #2
+
+declare <4 x i32> @llvm.arm.mve.orr.predicated.v4i32.v4i1(<4 x i32>, <4 x i32>, <4 x i1>, <4 x i32>) #2