[ARM][MVE][Intrinsics] Add *_x() variants of my *_m() intrinsics.

Summary:
Better use of multiclass is used, and this helped find some existing
bugs in the predicated VMULL* intrinsics, which are now fixed.

The refactored VMULL[TB]Q_(INT|POLY)_M() intrinsics were discovered
to have an argument ("inactive") with incorrect type, and this required
a fix that is included in this whole patch. The argument "inactive"
should have been the same width (per vector element) as the return
type of the intrinsic, but was not in the case where the return type
was double the element width of the input types.

To assist in testing the multiclassing , and to thwart further gremlins,
the unit tests are improved in scope.

The *.ll tests are all generated by a small bit of throw-away scripting
from the corresponding *.c tests, and as such the diffs are large and
nasty. Look at the file rather than the diff.

Reviewers: dmgreen, miyuki, ostannard, simon_tatham

Subscribers: kristof.beyls, hiraditya, cfe-commits, llvm-commits

Tags: #clang, #llvm

Differential Revision: https://reviews.llvm.org/D71421

19 files changed, 1301 insertions, 237 deletions

diff --git a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vabdq.ll b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vabdq.ll
index bafff00ea1d..fcd57664a16 100644
--- a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vabdq.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vabdq.ll
@@ -1,31 +1,61 @@
 ; 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 <4 x i32> @test_vabdq_u32(<4 x i32> %a, <4 x i32> %b) {
+define arm_aapcs_vfpcc <16 x i8> @test_vabdq_s8(<16 x i8> %a, <16 x i8> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vabdq_s8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vabd.s8 q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = tail call <16 x i8> @llvm.arm.mve.vabd.v16i8(<16 x i8> %a, <16 x i8> %b)
+  ret <16 x i8> %0
+}
+
+declare <16 x i8> @llvm.arm.mve.vabd.v16i8(<16 x i8>, <16 x i8>) #1
+
+define arm_aapcs_vfpcc <4 x i32> @test_vabdq_u32(<4 x i32> %a, <4 x i32> %b) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vabdq_u32:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vabd.s32 q0, q0, q1
 ; CHECK-NEXT:    bx lr
 entry:
-  %0 = tail call <4 x i32> @llvm.arm.mve.vabd.v4i32(<4 x i32>%a, <4 x i32>%b)
+  %0 = tail call <4 x i32> @llvm.arm.mve.vabd.v4i32(<4 x i32> %a, <4 x i32> %b)
   ret <4 x i32> %0
 }
 
-declare <4 x i32> @llvm.arm.mve.vabd.v4i32(<4 x i32>, <4 x i32>)
+declare <4 x i32> @llvm.arm.mve.vabd.v4i32(<4 x i32>, <4 x i32>) #1
 
-define arm_aapcs_vfpcc <4 x float> @test_vabdq_f32(<4 x float> %a, <4 x float> %b) {
+define arm_aapcs_vfpcc <8 x half> @test_vabdq_f32(<8 x half> %a, <8 x half> %b) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vabdq_f32:
 ; CHECK:       @ %bb.0: @ %entry
-; CHECK-NEXT:    vabd.f32 q0, q0, q1
+; CHECK-NEXT:    vabd.f16 q0, q0, q1
 ; CHECK-NEXT:    bx lr
 entry:
-  %0 = tail call <4 x float> @llvm.arm.mve.vabd.v4f32(<4 x float>%a, <4 x float>%b)
-  ret <4 x float> %0
+  %0 = tail call <8 x half> @llvm.arm.mve.vabd.v8f16(<8 x half> %a, <8 x half> %b)
+  ret <8 x half> %0
 }
 
-declare <4 x float> @llvm.arm.mve.vabd.v4f32(<4 x float>, <4 x float>)
+declare <8 x half> @llvm.arm.mve.vabd.v8f16(<8 x half>, <8 x half>) #1
 
-define arm_aapcs_vfpcc <16 x i8> @test_vabdq_m_s8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) {
+define arm_aapcs_vfpcc <8 x i16> @test_vabdq_m_u16(<8 x i16> %inactive, <8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vabdq_m_u16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vabdt.s16 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.abd.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) #1
+
+declare <8 x i16> @llvm.arm.mve.abd.predicated.v8i16.v8i1(<8 x i16>, <8 x i16>, <8 x i1>, <8 x i16>) #1
+
+define arm_aapcs_vfpcc <16 x i8> @test_vabdq_m_s8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vabdq_m_s8:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
@@ -39,24 +69,71 @@ entry:
   ret <16 x i8> %2
 }
 
-declare <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32)
+declare <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32) #1
+
+declare <16 x i8> @llvm.arm.mve.abd.predicated.v16i8.v16i1(<16 x i8>, <16 x i8>, <16 x i1>, <16 x i8>) #1
+
+define arm_aapcs_vfpcc <4 x float> @test_vabdq_m_f32(<4 x float> %inactive, <4 x float> %a, <4 x float> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vabdq_m_f32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vabdt.f32 q0, q1, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x float> @llvm.arm.mve.abd.predicated.v4f32.v4i1(<4 x float> %a, <4 x float> %b, <4 x i1> %1, <4 x float> %inactive)
+  ret <4 x float> %2
+}
+
+declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32) #1
+
+declare <4 x float> @llvm.arm.mve.abd.predicated.v4f32.v4i1(<4 x float>, <4 x float>, <4 x i1>, <4 x float>) #1
+
+define arm_aapcs_vfpcc <8 x i16> @test_vabdq_x_u16(<8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vabdq_x_u16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vabdt.s16 q0, q0, q1
+; 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.abd.predicated.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, <8 x i1> %1, <8 x i16> undef)
+  ret <8 x i16> %2
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vabdq_x_u32(<4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vabdq_x_u32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vabdt.s32 q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x i32> @llvm.arm.mve.abd.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> undef)
+  ret <4 x i32> %2
+}
 
-declare <16 x i8> @llvm.arm.mve.abd.predicated.v16i8.v16i1(<16 x i8>, <16 x i8>, <16 x i1>, <16 x i8>)
+declare <4 x i32> @llvm.arm.mve.abd.predicated.v4i32.v4i1(<4 x i32>, <4 x i32>, <4 x i1>, <4 x i32>) #1
 
-define arm_aapcs_vfpcc <8 x half> @test_vabdq_m_f16(<8 x half> %inactive, <8 x half> %a, <8 x half> %b, i16 zeroext %p) {
-; CHECK-LABEL: test_vabdq_m_f16:
+define arm_aapcs_vfpcc <8 x half> @test_vabdq_x_f16(<8 x half> %a, <8 x half> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vabdq_x_f16:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
 ; CHECK-NEXT:    vpst
-; CHECK-NEXT:    vabdt.f16 q0, q1, q2
+; CHECK-NEXT:    vabdt.f16 q0, q0, q1
 ; 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 half> @llvm.arm.mve.abd.predicated.v8f16.v8i1(<8 x half> %a, <8 x half> %b, <8 x i1> %1, <8 x half> %inactive)
+  %2 = tail call <8 x half> @llvm.arm.mve.abd.predicated.v8f16.v8i1(<8 x half> %a, <8 x half> %b, <8 x i1> %1, <8 x half> undef)
   ret <8 x half> %2
 }
 
-declare <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32)
+declare <8 x half> @llvm.arm.mve.abd.predicated.v8f16.v8i1(<8 x half>, <8 x half>, <8 x i1>, <8 x half>) #1
 
-declare <8 x half> @llvm.arm.mve.abd.predicated.v8f16.v8i1(<8 x half>, <8 x half>, <8 x i1>, <8 x half>)
diff --git a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vaddq.ll b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vaddq.ll
index a3cb91c21bc..b7becf6dd2c 100644
--- a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vaddq.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vaddq.ll
@@ -1,7 +1,7 @@
 ; 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 <4 x i32> @test_vaddq_u32(<4 x i32> %a, <4 x i32> %b) {
+define arm_aapcs_vfpcc <4 x i32> @test_vaddq_u32(<4 x i32> %a, <4 x i32> %b) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vaddq_u32:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vadd.i32 q0, q1, q0
@@ -11,37 +11,17 @@ entry:
   ret <4 x i32> %0
 }
 
-define arm_aapcs_vfpcc <4 x float> @test_vaddq_f32(<4 x float> %a, <4 x float> %b) {
-; CHECK-LABEL: test_vaddq_f32:
+define arm_aapcs_vfpcc <8 x half> @test_vaddq_f16(<8 x half> %a, <8 x half> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vaddq_f16:
 ; CHECK:       @ %bb.0: @ %entry
-; CHECK-NEXT:    vadd.f32 q0, q1, q0
+; CHECK-NEXT:    vadd.f16 q0, q0, q1
 ; CHECK-NEXT:    bx lr
 entry:
-  %0 = fadd <4 x float> %b, %a
-  ret <4 x float> %0
-}
-
-define arm_aapcs_vfpcc <8 x half> @test_vsubq_f16(<8 x half> %a, <8 x half> %b) {
-; CHECK-LABEL: test_vsubq_f16:
-; CHECK:       @ %bb.0: @ %entry
-; CHECK-NEXT:    vsub.f16 q0, q0, q1
-; CHECK-NEXT:    bx lr
-entry:
-  %0 = fsub <8 x half> %a, %b
+  %0 = fadd <8 x half> %a, %b
   ret <8 x half> %0
 }
 
-define arm_aapcs_vfpcc <8 x i16> @test_vsubq_s16(<8 x i16> %a, <8 x i16> %b) {
-; CHECK-LABEL: test_vsubq_s16:
-; CHECK:       @ %bb.0: @ %entry
-; CHECK-NEXT:    vsub.i16 q0, q0, q1
-; CHECK-NEXT:    bx lr
-entry:
-  %0 = sub <8 x i16> %a, %b
-  ret <8 x i16> %0
-}
-
-define arm_aapcs_vfpcc <16 x i8> @test_vaddq_m_s8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) {
+define arm_aapcs_vfpcc <16 x i8> @test_vaddq_m_s8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
 ; CHECK-LABEL: test_vaddq_m_s8:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
@@ -55,58 +35,59 @@ entry:
   ret <16 x i8> %2
 }
 
-declare <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32)
+declare <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32) #2
 
-declare <16 x i8> @llvm.arm.mve.add.predicated.v16i8.v16i1(<16 x i8>, <16 x i8>, <16 x i1>, <16 x i8>)
+declare <16 x i8> @llvm.arm.mve.add.predicated.v16i8.v16i1(<16 x i8>, <16 x i8>, <16 x i1>, <16 x i8>) #2
 
-define arm_aapcs_vfpcc <8 x half> @test_vaddq_m_f16(<8 x half> %inactive, <8 x half> %a, <8 x half> %b, i16 zeroext %p) {
-; CHECK-LABEL: test_vaddq_m_f16:
+define arm_aapcs_vfpcc <4 x float> @test_vaddq_m_f32(<4 x float> %inactive, <4 x float> %a, <4 x float> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vaddq_m_f32:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
 ; CHECK-NEXT:    vpst
-; CHECK-NEXT:    vaddt.f16 q0, q1, q2
+; CHECK-NEXT:    vaddt.f32 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 half> @llvm.arm.mve.add.predicated.v8f16.v8i1(<8 x half> %a, <8 x half> %b, <8 x i1> %1, <8 x half> %inactive)
-  ret <8 x half> %2
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x float> @llvm.arm.mve.add.predicated.v4f32.v4i1(<4 x float> %a, <4 x float> %b, <4 x i1> %1, <4 x float> %inactive)
+  ret <4 x float> %2
 }
 
-declare <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32)
+declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32) #2
 
-declare <8 x half> @llvm.arm.mve.add.predicated.v8f16.v8i1(<8 x half>, <8 x half>, <8 x i1>, <8 x half>)
+declare <4 x float> @llvm.arm.mve.add.predicated.v4f32.v4i1(<4 x float>, <4 x float>, <4 x i1>, <4 x float>) #2
 
-define arm_aapcs_vfpcc <4 x float> @test_vsubq_m_f32(<4 x float> %inactive, <4 x float> %a, <4 x float> %b, i16 zeroext %p) {
-; CHECK-LABEL: test_vsubq_m_f32:
+define arm_aapcs_vfpcc <8 x i16> @test_vaddq_x_u16(<8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vaddq_x_u16:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
 ; CHECK-NEXT:    vpst
-; CHECK-NEXT:    vsubt.f32 q0, q1, q2
+; CHECK-NEXT:    vaddt.i16 q0, q0, q1
 ; CHECK-NEXT:    bx lr
 entry:
   %0 = zext i16 %p to i32
-  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
-  %2 = tail call <4 x float> @llvm.arm.mve.sub.predicated.v4f32.v4i1(<4 x float> %a, <4 x float> %b, <4 x i1> %1, <4 x float> %inactive)
-  ret <4 x float> %2
+  %1 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %0)
+  %2 = tail call <8 x i16> @llvm.arm.mve.add.predicated.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, <8 x i1> %1, <8 x i16> undef)
+  ret <8 x i16> %2
 }
 
-declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32)
+declare <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32) #2
 
-declare <4 x float> @llvm.arm.mve.sub.predicated.v4f32.v4i1(<4 x float>, <4 x float>, <4 x i1>, <4 x float>)
+declare <8 x i16> @llvm.arm.mve.add.predicated.v8i16.v8i1(<8 x i16>, <8 x i16>, <8 x i1>, <8 x i16>) #2
 
-define arm_aapcs_vfpcc <4 x i32> @test_vsubq_m_u32(<4 x i32> %inactive, <4 x i32> %a, <4 x i32> %b, i16 zeroext %p) {
-; CHECK-LABEL: test_vsubq_m_u32:
+define arm_aapcs_vfpcc <8 x half> @test_vaddq_x_f16(<8 x half> %a, <8 x half> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vaddq_x_f16:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
 ; CHECK-NEXT:    vpst
-; CHECK-NEXT:    vsubt.i32 q0, q1, q2
+; CHECK-NEXT:    vaddt.f16 q0, q0, q1
 ; CHECK-NEXT:    bx lr
 entry:
   %0 = zext i16 %p to i32
-  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
-  %2 = tail call <4 x i32> @llvm.arm.mve.sub.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> %inactive)
-  ret <4 x i32> %2
+  %1 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %0)
+  %2 = tail call <8 x half> @llvm.arm.mve.add.predicated.v8f16.v8i1(<8 x half> %a, <8 x half> %b, <8 x i1> %1, <8 x half> undef)
+  ret <8 x half> %2
 }
 
-declare <4 x i32> @llvm.arm.mve.sub.predicated.v4i32.v4i1(<4 x i32>, <4 x i32>, <4 x i1>, <4 x i32>)
+declare <8 x half> @llvm.arm.mve.add.predicated.v8f16.v8i1(<8 x half>, <8 x half>, <8 x i1>, <8 x half>) #2
+
diff --git a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vandq.ll b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vandq.ll
index 1b1d498bc37..651ac17234b 100644
--- a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vandq.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vandq.ll
@@ -11,24 +11,24 @@ entry:
   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:
+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 <4 x i32> %b, %a
-  ret <4 x i32> %0
+  %0 = and <8 x i16> %b, %a
+  ret <8 x i16> %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:
+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 <8 x i16> %b, %a
-  ret <8 x i16> %0
+  %0 = and <4 x i32> %b, %a
+  ret <4 x i32> %0
 }
 
 define arm_aapcs_vfpcc <4 x float> @test_vandq_f32(<4 x float> %a, <4 x float> %b) local_unnamed_addr #0 {
@@ -80,25 +80,98 @@ 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:
+define arm_aapcs_vfpcc <4 x i32> @test_vandq_m_s32(<4 x i32> %inactive, <4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vandq_m_s32:
+; 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 <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x i32> @llvm.arm.mve.and.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> %inactive)
+  ret <4 x i32> %2
+}
+
+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
+
+define arm_aapcs_vfpcc <8 x half> @test_vandq_m_f16(<8 x half> %inactive, <8 x half> %a, <8 x half> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vandq_m_f16:
 ; 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 <8 x half> %a to <8 x i16>
+  %1 = bitcast <8 x half> %b to <8 x i16>
+  %2 = zext i16 %p to i32
+  %3 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %2)
+  %4 = bitcast <8 x half> %inactive to <8 x i16>
+  %5 = tail call <8 x i16> @llvm.arm.mve.and.predicated.v8i16.v8i1(<8 x i16> %0, <8 x i16> %1, <8 x i1> %3, <8 x i16> %4)
+  %6 = bitcast <8 x i16> %5 to <8 x half>
+  ret <8 x half> %6
+}
+
+define arm_aapcs_vfpcc <16 x i8> @test_vandq_x_u8(<16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vandq_x_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vandt q0, q0, q1
+; 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> undef)
+  ret <16 x i8> %2
+}
+
+define arm_aapcs_vfpcc <8 x i16> @test_vandq_x_s16(<8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vandq_x_s16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vandt q0, q0, q1
+; 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> undef)
+  ret <8 x i16> %2
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vandq_x_u32(<4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vandq_x_u32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vandt q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x i32> @llvm.arm.mve.and.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> undef)
+  ret <4 x i32> %2
+}
+
+define arm_aapcs_vfpcc <4 x float> @test_vandq_m_f32(<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, 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 = 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
+  %4 = 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> undef)
+  %5 = bitcast <4 x i32> %4 to <4 x float>
+  ret <4 x float> %5
 }
 
-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
diff --git a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vbicq.ll b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vbicq.ll
index 47877a13cb9..97b7ac9fc34 100644
--- a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vbicq.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vbicq.ll
@@ -12,26 +12,26 @@ entry:
   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:
+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 <4 x i32> %b, <i32 -1, i32 -1, i32 -1, i32 -1>
-  %1 = and <4 x i32> %0, %a
-  ret <4 x i32> %1
+  %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 <8 x i16> @test_vbicq_s16(<8 x i16> %a, <8 x i16> %b) local_unnamed_addr #0 {
-; CHECK-LABEL: test_vbicq_s16:
+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 <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
+  %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 <4 x float> @test_vbicq_f32(<4 x float> %a, <4 x float> %b) local_unnamed_addr #0 {
@@ -84,25 +84,98 @@ 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:
+define arm_aapcs_vfpcc <4 x i32> @test_vbicq_m_s32(<4 x i32> %inactive, <4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vbicq_m_s32:
 ; 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 <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x i32> @llvm.arm.mve.bic.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> %inactive)
+  ret <4 x i32> %2
+}
+
+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
+
+define arm_aapcs_vfpcc <8 x half> @test_vbicq_m_f16(<8 x half> %inactive, <8 x half> %a, <8 x half> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vbicq_m_f16:
+; 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 <8 x half> %a to <8 x i16>
+  %1 = bitcast <8 x half> %b to <8 x i16>
+  %2 = zext i16 %p to i32
+  %3 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %2)
+  %4 = bitcast <8 x half> %inactive to <8 x i16>
+  %5 = tail call <8 x i16> @llvm.arm.mve.bic.predicated.v8i16.v8i1(<8 x i16> %0, <8 x i16> %1, <8 x i1> %3, <8 x i16> %4)
+  %6 = bitcast <8 x i16> %5 to <8 x half>
+  ret <8 x half> %6
+}
+
+define arm_aapcs_vfpcc <16 x i8> @test_vbicq_x_u8(<16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vbicq_x_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vbict q0, q0, q1
+; 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> undef)
+  ret <16 x i8> %2
+}
+
+define arm_aapcs_vfpcc <8 x i16> @test_vbicq_x_s16(<8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vbicq_x_s16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vbict q0, q0, q1
+; 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> undef)
+  ret <8 x i16> %2
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vbicq_x_u32(<4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vbicq_x_u32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vbict q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x i32> @llvm.arm.mve.bic.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> undef)
+  ret <4 x i32> %2
+}
+
+define arm_aapcs_vfpcc <4 x float> @test_vbicq_m_f32(<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, 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 = 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
+  %4 = 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> undef)
+  %5 = bitcast <4 x i32> %4 to <4 x float>
+  ret <4 x float> %5
 }
 
-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
diff --git a/llvm/test/CodeGen/Thumb2/mve-intrinsics/veorq.ll b/llvm/test/CodeGen/Thumb2/mve-intrinsics/veorq.ll
index 9b66f3656eb..6f70cafb013 100644
--- a/llvm/test/CodeGen/Thumb2/mve-intrinsics/veorq.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-intrinsics/veorq.ll
@@ -11,24 +11,24 @@ entry:
   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:
+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 <4 x i32> %b, %a
-  ret <4 x i32> %0
+  %0 = xor <8 x i16> %b, %a
+  ret <8 x i16> %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:
+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 <8 x i16> %b, %a
-  ret <8 x i16> %0
+  %0 = xor <4 x i32> %b, %a
+  ret <4 x i32> %0
 }
 
 define arm_aapcs_vfpcc <4 x float> @test_veorq_f32(<4 x float> %a, <4 x float> %b) local_unnamed_addr #0 {
@@ -80,25 +80,98 @@ 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:
+define arm_aapcs_vfpcc <4 x i32> @test_veorq_m_s32(<4 x i32> %inactive, <4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_veorq_m_s32:
+; 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 <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x i32> @llvm.arm.mve.eor.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> %inactive)
+  ret <4 x i32> %2
+}
+
+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
+
+define arm_aapcs_vfpcc <8 x half> @test_veorq_m_f16(<8 x half> %inactive, <8 x half> %a, <8 x half> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_veorq_m_f16:
 ; 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 <8 x half> %a to <8 x i16>
+  %1 = bitcast <8 x half> %b to <8 x i16>
+  %2 = zext i16 %p to i32
+  %3 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %2)
+  %4 = bitcast <8 x half> %inactive to <8 x i16>
+  %5 = tail call <8 x i16> @llvm.arm.mve.eor.predicated.v8i16.v8i1(<8 x i16> %0, <8 x i16> %1, <8 x i1> %3, <8 x i16> %4)
+  %6 = bitcast <8 x i16> %5 to <8 x half>
+  ret <8 x half> %6
+}
+
+define arm_aapcs_vfpcc <16 x i8> @test_veorq_x_u8(<16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_veorq_x_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    veort q0, q0, q1
+; 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> undef)
+  ret <16 x i8> %2
+}
+
+define arm_aapcs_vfpcc <8 x i16> @test_veorq_x_s16(<8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_veorq_x_s16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    veort q0, q0, q1
+; 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> undef)
+  ret <8 x i16> %2
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_veorq_x_u32(<4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_veorq_x_u32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    veort q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x i32> @llvm.arm.mve.eor.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> undef)
+  ret <4 x i32> %2
+}
+
+define arm_aapcs_vfpcc <4 x float> @test_veorq_m_f32(<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, 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 = 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
+  %4 = 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> undef)
+  %5 = bitcast <4 x i32> %4 to <4 x float>
+  ret <4 x float> %5
 }
 
-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
diff --git a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vhaddq.ll b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vhaddq.ll
index 11b21142427..82ecc062c08 100644
--- a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vhaddq.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vhaddq.ll
@@ -90,3 +90,46 @@ entry:
 declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32) #1
 
 declare <4 x i32> @llvm.arm.mve.hadd.predicated.v4i32.v4i1(<4 x i32>, <4 x i32>, <4 x i1>, <4 x i32>) #1
+
+define arm_aapcs_vfpcc <16 x i8> @test_vhaddq_x_u8(<16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vhaddq_x_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vhaddt.s8 q0, q0, q1
+; 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.hadd.predicated.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, <16 x i1> %1, <16 x i8> undef)
+  ret <16 x i8> %2
+}
+
+define arm_aapcs_vfpcc <8 x i16> @test_vhaddq_x_s16(<8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vhaddq_x_s16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vhaddt.s16 q0, q0, q1
+; 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.hadd.predicated.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, <8 x i1> %1, <8 x i16> undef)
+  ret <8 x i16> %2
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vhaddq_x_u32(<4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vhaddq_x_u32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vhaddt.s32 q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x i32> @llvm.arm.mve.hadd.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> undef)
+  ret <4 x i32> %2
+}
+
diff --git a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmaxnmq.ll b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmaxnmq.ll
index d89308bb594..54a140042bf 100644
--- a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmaxnmq.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmaxnmq.ll
@@ -1,7 +1,7 @@
 ; 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 dso_local arm_aapcs_vfpcc <8 x half> @test_vmaxnmq_f16(<8 x half> %a, <8 x half> %b) local_unnamed_addr #0 {
+define arm_aapcs_vfpcc <8 x half> @test_vmaxnmq_f16(<8 x half> %a, <8 x half> %b) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vmaxnmq_f16:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmaxnm.f16 q0, q0, q1
@@ -13,7 +13,7 @@ entry:
 
 declare <8 x half> @llvm.maxnum.v8f16(<8 x half>, <8 x half>) #1
 
-define dso_local arm_aapcs_vfpcc <4 x float> @test_vmaxnmq_f32(<4 x float> %a, <4 x float> %b) local_unnamed_addr #0 {
+define arm_aapcs_vfpcc <4 x float> @test_vmaxnmq_f32(<4 x float> %a, <4 x float> %b) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vmaxnmq_f32:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmaxnm.f32 q0, q0, q1
@@ -25,7 +25,7 @@ entry:
 
 declare <4 x float> @llvm.maxnum.v4f32(<4 x float>, <4 x float>) #1
 
-define dso_local arm_aapcs_vfpcc <8 x half> @test_vmaxnmq_m_f16(<8 x half> %inactive, <8 x half> %a, <8 x half> %b, i16 zeroext %p) local_unnamed_addr #0 {
+define arm_aapcs_vfpcc <8 x half> @test_vmaxnmq_m_f16(<8 x half> %inactive, <8 x half> %a, <8 x half> %b, i16 zeroext %p) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vmaxnmq_m_f16:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
@@ -38,11 +38,12 @@ entry:
   %2 = tail call <8 x half> @llvm.arm.mve.max.predicated.v8f16.v8i1(<8 x half> %a, <8 x half> %b, <8 x i1> %1, <8 x half> %inactive)
   ret <8 x half> %2
 }
+
 declare <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32) #2
 
 declare <8 x half> @llvm.arm.mve.max.predicated.v8f16.v8i1(<8 x half>, <8 x half>, <8 x i1>, <8 x half>) #2
 
-define dso_local arm_aapcs_vfpcc <4 x float> @test_vmaxnmq_m_f32(<4 x float> %inactive, <4 x float> %a, <4 x float> %b, i16 zeroext %p) local_unnamed_addr #0 {
+define arm_aapcs_vfpcc <4 x float> @test_vmaxnmq_m_f32(<4 x float> %inactive, <4 x float> %a, <4 x float> %b, i16 zeroext %p) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vmaxnmq_m_f32:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
@@ -59,3 +60,32 @@ entry:
 declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32) #2
 
 declare <4 x float> @llvm.arm.mve.max.predicated.v4f32.v4i1(<4 x float>, <4 x float>, <4 x i1>, <4 x float>) #2
+
+define arm_aapcs_vfpcc <8 x half> @test_vmaxnmq_x_f16(<8 x half> %a, <8 x half> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vmaxnmq_x_f16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmaxnmt.f32 q0, q0, q1
+; 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 half> @llvm.arm.mve.max.predicated.v8f16.v8i1(<8 x half> %a, <8 x half> %b, <8 x i1> %1, <8 x half> undef)
+  ret <8 x half> %2
+}
+
+define arm_aapcs_vfpcc <4 x float> @test_vmaxnmq_x_f32(<4 x float> %a, <4 x float> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vmaxnmq_x_f32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmaxnmt.f32 q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x float> @llvm.arm.mve.max.predicated.v4f32.v4i1(<4 x float> %a, <4 x float> %b, <4 x i1> %1, <4 x float> undef)
+  ret <4 x float> %2
+}
+
diff --git a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmaxq.ll b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmaxq.ll
index 09a7d60cd16..622aa1ccd33 100644
--- a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmaxq.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmaxq.ll
@@ -1,89 +1,132 @@
 ; 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 dso_local arm_aapcs_vfpcc <16 x i8> @test_vmaxq_u8(<16 x i8> %a, <16 x i8> %b) local_unnamed_addr #0 {
-; CHECK-LABEL: test_vmaxq_u8:
+define arm_aapcs_vfpcc <16 x i8> @test_vmaxq_s8(<16 x i8> %a, <16 x i8> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vmaxq_s8:
 ; CHECK:       @ %bb.0: @ %entry
-; CHECK-NEXT:    vmin.u8 q0, q0, q1
+; CHECK-NEXT:    vmax.s8 q0, q0, q1
 ; CHECK-NEXT:    bx lr
 entry:
-  %0 = icmp ugt <16 x i8> %a, %b
+  %0 = icmp slt <16 x i8> %a, %b
   %1 = select <16 x i1> %0, <16 x i8> %b, <16 x i8> %a
   ret <16 x i8> %1
 }
 
-define dso_local arm_aapcs_vfpcc <8 x i16> @test_vmaxq_s16(<8 x i16> %a, <8 x i16> %b) local_unnamed_addr #0 {
-; CHECK-LABEL: test_vmaxq_s16:
+define arm_aapcs_vfpcc <8 x i16> @test_vmaxq_u16(<8 x i16> %a, <8 x i16> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vmaxq_u16:
 ; CHECK:       @ %bb.0: @ %entry
-; CHECK-NEXT:    vmin.s16 q0, q0, q1
+; CHECK-NEXT:    vmax.u16 q0, q0, q1
 ; CHECK-NEXT:    bx lr
 entry:
-  %0 = icmp sgt <8 x i16> %a, %b
+  %0 = icmp ult <8 x i16> %a, %b
   %1 = select <8 x i1> %0, <8 x i16> %b, <8 x i16> %a
   ret <8 x i16> %1
 }
 
-define dso_local arm_aapcs_vfpcc <4 x i32> @test_vmaxq_u32(<4 x i32> %a, <4 x i32> %b) local_unnamed_addr #0 {
-; CHECK-LABEL: test_vmaxq_u32:
+define arm_aapcs_vfpcc <4 x i32> @test_vmaxq_s32(<4 x i32> %a, <4 x i32> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vmaxq_s32:
 ; CHECK:       @ %bb.0: @ %entry
-; CHECK-NEXT:    vmin.u32 q0, q0, q1
+; CHECK-NEXT:    vmax.s32 q0, q0, q1
 ; CHECK-NEXT:    bx lr
 entry:
-  %0 = icmp ugt <4 x i32> %a, %b
+  %0 = icmp slt <4 x i32> %a, %b
   %1 = select <4 x i1> %0, <4 x i32> %b, <4 x i32> %a
   ret <4 x i32> %1
 }
 
-define dso_local arm_aapcs_vfpcc <16 x i8> @test_vmaxq_m_s8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
-; CHECK-LABEL: test_vmaxq_m_s8:
+define arm_aapcs_vfpcc <16 x i8> @test_vmaxq_m_u8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vmaxq_m_u8:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
 ; CHECK-NEXT:    vpst
-; CHECK-NEXT:    vmint.s8 q0, q1, q2
+; CHECK-NEXT:    vmaxt.s8 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.min.predicated.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, <16 x i1> %1, <16 x i8> %inactive)
+  %2 = tail call <16 x i8> @llvm.arm.mve.max.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.min.predicated.v16i8.v16i1(<16 x i8>, <16 x i8>, <16 x i1>, <16 x i8>) #2
+declare <16 x i8> @llvm.arm.mve.max.predicated.v16i8.v16i1(<16 x i8>, <16 x i8>, <16 x i1>, <16 x i8>) #2
 
-define dso_local arm_aapcs_vfpcc <8 x i16> @test_vmaxq_m_u16(<8 x i16> %inactive, <8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
-; CHECK-LABEL: test_vmaxq_m_u16:
+define arm_aapcs_vfpcc <8 x i16> @test_vmaxq_m_s16(<8 x i16> %inactive, <8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vmaxq_m_s16:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
 ; CHECK-NEXT:    vpst
-; CHECK-NEXT:    vmint.s16 q0, q1, q2
+; CHECK-NEXT:    vmaxt.s16 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.min.predicated.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, <8 x i1> %1, <8 x i16> %inactive)
+  %2 = tail call <8 x i16> @llvm.arm.mve.max.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.min.predicated.v8i16.v8i1(<8 x i16>, <8 x i16>, <8 x i1>, <8 x i16>) #2
+declare <8 x i16> @llvm.arm.mve.max.predicated.v8i16.v8i1(<8 x i16>, <8 x i16>, <8 x i1>, <8 x i16>) #2
 
-define dso_local arm_aapcs_vfpcc <4 x i32> @test_vmaxq_m_s32(<4 x i32> %inactive, <4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #1 {
-; CHECK-LABEL: test_vmaxq_m_s32:
+define arm_aapcs_vfpcc <4 x i32> @test_vmaxq_m_u32(<4 x i32> %inactive, <4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vmaxq_m_u32:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
 ; CHECK-NEXT:    vpst
-; CHECK-NEXT:    vmint.s32 q0, q1, q2
+; CHECK-NEXT:    vmaxt.s32 q0, q1, q2
 ; CHECK-NEXT:    bx lr
 entry:
   %0 = zext i16 %p to i32
   %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
-  %2 = tail call <4 x i32> @llvm.arm.mve.min.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> %inactive)
+  %2 = tail call <4 x i32> @llvm.arm.mve.max.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> %inactive)
   ret <4 x i32> %2
 }
 
 declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32) #2
 
-declare <4 x i32> @llvm.arm.mve.min.predicated.v4i32.v4i1(<4 x i32>, <4 x i32>, <4 x i1>, <4 x i32>) #2
+declare <4 x i32> @llvm.arm.mve.max.predicated.v4i32.v4i1(<4 x i32>, <4 x i32>, <4 x i1>, <4 x i32>) #2
+
+define arm_aapcs_vfpcc <16 x i8> @test_vmaxq_x_u8(<16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vmaxq_x_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmaxt.s8 q0, q0, q1
+; 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.max.predicated.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, <16 x i1> %1, <16 x i8> undef)
+  ret <16 x i8> %2
+}
+
+define arm_aapcs_vfpcc <8 x i16> @test_vmaxq_x_u16(<8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vmaxq_x_u16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmaxt.s16 q0, q0, q1
+; 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.max.predicated.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, <8 x i1> %1, <8 x i16> undef)
+  ret <8 x i16> %2
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vmaxq_x_s32(<4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vmaxq_x_s32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmaxt.s32 q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x i32> @llvm.arm.mve.max.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> undef)
+  ret <4 x i32> %2
+}
+
diff --git a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vminnmq.ll b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vminnmq.ll
index 10cd674d39a..ae449099523 100644
--- a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vminnmq.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vminnmq.ll
@@ -1,7 +1,7 @@
 ; 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 dso_local arm_aapcs_vfpcc <8 x half> @test_vminnmq_f16(<8 x half> %a, <8 x half> %b) local_unnamed_addr #0 {
+define arm_aapcs_vfpcc <8 x half> @test_vminnmq_f16(<8 x half> %a, <8 x half> %b) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vminnmq_f16:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vminnm.f16 q0, q0, q1
@@ -13,7 +13,7 @@ entry:
 
 declare <8 x half> @llvm.minnum.v8f16(<8 x half>, <8 x half>) #1
 
-define dso_local arm_aapcs_vfpcc <4 x float> @test_vminnmq_f32(<4 x float> %a, <4 x float> %b) local_unnamed_addr #0 {
+define arm_aapcs_vfpcc <4 x float> @test_vminnmq_f32(<4 x float> %a, <4 x float> %b) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vminnmq_f32:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vminnm.f32 q0, q0, q1
@@ -25,7 +25,7 @@ entry:
 
 declare <4 x float> @llvm.minnum.v4f32(<4 x float>, <4 x float>) #1
 
-define dso_local arm_aapcs_vfpcc <8 x half> @test_vminnmq_m_f16(<8 x half> %inactive, <8 x half> %a, <8 x half> %b, i16 zeroext %p) local_unnamed_addr #0 {
+define arm_aapcs_vfpcc <8 x half> @test_vminnmq_m_f16(<8 x half> %inactive, <8 x half> %a, <8 x half> %b, i16 zeroext %p) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vminnmq_m_f16:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
@@ -43,7 +43,7 @@ declare <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32) #2
 
 declare <8 x half> @llvm.arm.mve.min.predicated.v8f16.v8i1(<8 x half>, <8 x half>, <8 x i1>, <8 x half>) #2
 
-define dso_local arm_aapcs_vfpcc <4 x float> @test_vminnmq_m_f32(<4 x float> %inactive, <4 x float> %a, <4 x float> %b, i16 zeroext %p) local_unnamed_addr #0 {
+define arm_aapcs_vfpcc <4 x float> @test_vminnmq_m_f32(<4 x float> %inactive, <4 x float> %a, <4 x float> %b, i16 zeroext %p) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vminnmq_m_f32:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
@@ -60,3 +60,32 @@ entry:
 declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32) #2
 
 declare <4 x float> @llvm.arm.mve.min.predicated.v4f32.v4i1(<4 x float>, <4 x float>, <4 x i1>, <4 x float>) #2
+
+define arm_aapcs_vfpcc <8 x half> @test_vminnmq_x_f16(<8 x half> %a, <8 x half> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vminnmq_x_f16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vminnmt.f32 q0, q0, q1
+; 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 half> @llvm.arm.mve.min.predicated.v8f16.v8i1(<8 x half> %a, <8 x half> %b, <8 x i1> %1, <8 x half> undef)
+  ret <8 x half> %2
+}
+
+define arm_aapcs_vfpcc <4 x float> @test_vminnmq_x_f32(<4 x float> %a, <4 x float> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vminnmq_x_f32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vminnmt.f32 q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x float> @llvm.arm.mve.min.predicated.v4f32.v4i1(<4 x float> %a, <4 x float> %b, <4 x i1> %1, <4 x float> undef)
+  ret <4 x float> %2
+}
+
diff --git a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vminq.ll b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vminq.ll
index 0cbef86c928..96bc2233f22 100644
--- a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vminq.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vminq.ll
@@ -1,7 +1,7 @@
 ; 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 dso_local arm_aapcs_vfpcc <16 x i8> @test_vminq_u8(<16 x i8> %a, <16 x i8> %b) local_unnamed_addr #0 {
+define arm_aapcs_vfpcc <16 x i8> @test_vminq_u8(<16 x i8> %a, <16 x i8> %b) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vminq_u8:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmin.u8 q0, q0, q1
@@ -12,7 +12,7 @@ entry:
   ret <16 x i8> %1
 }
 
-define dso_local arm_aapcs_vfpcc <8 x i16> @test_vminq_s16(<8 x i16> %a, <8 x i16> %b) local_unnamed_addr #0 {
+define arm_aapcs_vfpcc <8 x i16> @test_vminq_s16(<8 x i16> %a, <8 x i16> %b) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vminq_s16:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmin.s16 q0, q0, q1
@@ -23,7 +23,7 @@ entry:
   ret <8 x i16> %1
 }
 
-define dso_local arm_aapcs_vfpcc <4 x i32> @test_vminq_u32(<4 x i32> %a, <4 x i32> %b) local_unnamed_addr #0 {
+define arm_aapcs_vfpcc <4 x i32> @test_vminq_u32(<4 x i32> %a, <4 x i32> %b) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vminq_u32:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmin.u32 q0, q0, q1
@@ -34,7 +34,7 @@ entry:
   ret <4 x i32> %1
 }
 
-define dso_local arm_aapcs_vfpcc <16 x i8> @test_vminq_m_s8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
+define arm_aapcs_vfpcc <16 x i8> @test_vminq_m_s8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
 ; CHECK-LABEL: test_vminq_m_s8:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
@@ -52,7 +52,7 @@ declare <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32) #2
 
 declare <16 x i8> @llvm.arm.mve.min.predicated.v16i8.v16i1(<16 x i8>, <16 x i8>, <16 x i1>, <16 x i8>) #2
 
-define dso_local arm_aapcs_vfpcc <8 x i16> @test_vminq_m_u16(<8 x i16> %inactive, <8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
+define arm_aapcs_vfpcc <8 x i16> @test_vminq_m_u16(<8 x i16> %inactive, <8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
 ; CHECK-LABEL: test_vminq_m_u16:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
@@ -70,7 +70,7 @@ declare <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32) #2
 
 declare <8 x i16> @llvm.arm.mve.min.predicated.v8i16.v8i1(<8 x i16>, <8 x i16>, <8 x i1>, <8 x i16>) #2
 
-define dso_local arm_aapcs_vfpcc <4 x i32> @test_vminq_m_s32(<4 x i32> %inactive, <4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #1 {
+define arm_aapcs_vfpcc <4 x i32> @test_vminq_m_s32(<4 x i32> %inactive, <4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #1 {
 ; CHECK-LABEL: test_vminq_m_s32:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
@@ -87,3 +87,46 @@ entry:
 declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32) #2
 
 declare <4 x i32> @llvm.arm.mve.min.predicated.v4i32.v4i1(<4 x i32>, <4 x i32>, <4 x i1>, <4 x i32>) #2
+
+define arm_aapcs_vfpcc <16 x i8> @test_vminq_x_u8(<16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vminq_x_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmint.s8 q0, q0, q1
+; 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.min.predicated.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, <16 x i1> %1, <16 x i8> undef)
+  ret <16 x i8> %2
+}
+
+define arm_aapcs_vfpcc <8 x i16> @test_vminq_x_s16(<8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vminq_x_s16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmint.s16 q0, q0, q1
+; 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.min.predicated.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, <8 x i1> %1, <8 x i16> undef)
+  ret <8 x i16> %2
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vminq_x_s32(<4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vminq_x_s32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmint.s32 q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x i32> @llvm.arm.mve.min.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> undef)
+  ret <4 x i32> %2
+}
+
diff --git a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmulhq.ll b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmulhq.ll
index 78ee17b5541..ba15af6c549 100644
--- a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmulhq.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmulhq.ll
@@ -90,3 +90,46 @@ entry:
 declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32) #1
 
 declare <4 x i32> @llvm.arm.mve.mulh.predicated.v4i32.v4i1(<4 x i32>, <4 x i32>, <4 x i1>, <4 x i32>) #1
+
+define arm_aapcs_vfpcc <16 x i8> @test_vmulhq_x_u8(<16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vmulhq_x_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmulht.s8 q0, q0, q1
+; 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.mulh.predicated.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, <16 x i1> %1, <16 x i8> undef)
+  ret <16 x i8> %2
+}
+
+define arm_aapcs_vfpcc <8 x i16> @test_vmulhq_x_s16(<8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vmulhq_x_s16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmulht.s16 q0, q0, q1
+; 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.mulh.predicated.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, <8 x i1> %1, <8 x i16> undef)
+  ret <8 x i16> %2
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vmulhq_x_u32(<4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vmulhq_x_u32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmulht.s32 q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x i32> @llvm.arm.mve.mulh.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> undef)
+  ret <4 x i32> %2
+}
+
diff --git a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmullbq.ll b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmullbq.ll
index 78f24fa62c2..e443d54f020 100644
--- a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmullbq.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmullbq.ll
@@ -50,7 +50,7 @@ entry:
 
 declare <4 x i32> @llvm.arm.mve.vmull.poly.v4i32.v8i16(<8 x i16>, <8 x i16>, i32) #1
 
-define arm_aapcs_vfpcc <8 x i16> @test_vmullbq_int_m_s8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #0 {
+define arm_aapcs_vfpcc <8 x i16> @test_vmullbq_int_m_s8(<8 x i16> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vmullbq_int_m_s8:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
@@ -60,15 +60,15 @@ define arm_aapcs_vfpcc <8 x i16> @test_vmullbq_int_m_s8(<16 x i8> %inactive, <16
 entry:
   %0 = zext i16 %p to i32
   %1 = tail call <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32 %0)
-  %2 = tail call <8 x i16> @llvm.arm.mve.mull.int.predicated.v8i16.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, i32 0, <16 x i1> %1, <16 x i8> %inactive)
+  %2 = tail call <8 x i16> @llvm.arm.mve.mull.int.predicated.v8i16.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, i32 0, <16 x i1> %1, <8 x i16> %inactive)
   ret <8 x i16> %2
 }
 
 declare <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32) #1
 
-declare <8 x i16> @llvm.arm.mve.mull.int.predicated.v8i16.v16i8.v16i1(<16 x i8>, <16 x i8>, i32, <16 x i1>, <16 x i8>) #1
+declare <8 x i16> @llvm.arm.mve.mull.int.predicated.v8i16.v16i8.v16i1(<16 x i8>, <16 x i8>, i32, <16 x i1>, <8 x i16>) #1
 
-define arm_aapcs_vfpcc <4 x i32> @test_vmullbq_int_m_u16(<8 x i16> %inactive, <8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #0 {
+define arm_aapcs_vfpcc <4 x i32> @test_vmullbq_int_m_u16(<4 x i32> %inactive, <8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vmullbq_int_m_u16:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
@@ -78,15 +78,15 @@ define arm_aapcs_vfpcc <4 x i32> @test_vmullbq_int_m_u16(<8 x i16> %inactive, <8
 entry:
   %0 = zext i16 %p to i32
   %1 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %0)
-  %2 = tail call <4 x i32> @llvm.arm.mve.mull.int.predicated.v4i32.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, i32 0, <8 x i1> %1, <8 x i16> %inactive)
+  %2 = tail call <4 x i32> @llvm.arm.mve.mull.int.predicated.v4i32.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, i32 0, <8 x i1> %1, <4 x i32> %inactive)
   ret <4 x i32> %2
 }
 
 declare <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32) #1
 
-declare <4 x i32> @llvm.arm.mve.mull.int.predicated.v4i32.v8i16.v8i1(<8 x i16>, <8 x i16>, i32, <8 x i1>, <8 x i16>) #1
+declare <4 x i32> @llvm.arm.mve.mull.int.predicated.v4i32.v8i16.v8i1(<8 x i16>, <8 x i16>, i32, <8 x i1>, <4 x i32>) #1
 
-define arm_aapcs_vfpcc <2 x i64> @test_vmullbq_int_m_s32(<4 x i32> %inactive, <4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #0 {
+define arm_aapcs_vfpcc <2 x i64> @test_vmullbq_int_m_s32(<2 x i64> %inactive, <4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vmullbq_int_m_s32:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
@@ -96,15 +96,15 @@ define arm_aapcs_vfpcc <2 x i64> @test_vmullbq_int_m_s32(<4 x i32> %inactive, <4
 entry:
   %0 = zext i16 %p to i32
   %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
-  %2 = tail call <2 x i64> @llvm.arm.mve.mull.int.predicated.v2i64.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, i32 0, <4 x i1> %1, <4 x i32> %inactive)
+  %2 = tail call <2 x i64> @llvm.arm.mve.mull.int.predicated.v2i64.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, i32 0, <4 x i1> %1, <2 x i64> %inactive)
   ret <2 x i64> %2
 }
 
 declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32) #1
 
-declare <2 x i64> @llvm.arm.mve.mull.int.predicated.v2i64.v4i32.v4i1(<4 x i32>, <4 x i32>, i32, <4 x i1>, <4 x i32>) #1
+declare <2 x i64> @llvm.arm.mve.mull.int.predicated.v2i64.v4i32.v4i1(<4 x i32>, <4 x i32>, i32, <4 x i1>, <2 x i64>) #1
 
-define arm_aapcs_vfpcc <8 x i16> @test_vmullbq_poly_m_p8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #0 {
+define arm_aapcs_vfpcc <8 x i16> @test_vmullbq_poly_m_p8(<8 x i16> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vmullbq_poly_m_p8:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
@@ -114,8 +114,68 @@ define arm_aapcs_vfpcc <8 x i16> @test_vmullbq_poly_m_p8(<16 x i8> %inactive, <1
 entry:
   %0 = zext i16 %p to i32
   %1 = tail call <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32 %0)
-  %2 = tail call <8 x i16> @llvm.arm.mve.mull.poly.predicated.v8i16.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, i32 0, <16 x i1> %1, <16 x i8> %inactive)
+  %2 = tail call <8 x i16> @llvm.arm.mve.mull.poly.predicated.v8i16.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, i32 0, <16 x i1> %1, <8 x i16> %inactive)
   ret <8 x i16> %2
 }
 
-declare <8 x i16> @llvm.arm.mve.mull.poly.predicated.v8i16.v16i8.v16i1(<16 x i8>, <16 x i8>, i32, <16 x i1>, <16 x i8>) #1
+declare <8 x i16> @llvm.arm.mve.mull.poly.predicated.v8i16.v16i8.v16i1(<16 x i8>, <16 x i8>, i32, <16 x i1>, <8 x i16>) #1
+
+define arm_aapcs_vfpcc <8 x i16> @test_vmullbq_int_x_u8(<16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vmullbq_int_x_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmullbt.s8 q0, q0, q1
+; 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 <8 x i16> @llvm.arm.mve.mull.int.predicated.v8i16.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, i32 0, <16 x i1> %1, <8 x i16> undef)
+  ret <8 x i16> %2
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vmullbq_int_x_s16(<8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vmullbq_int_x_s16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmullbt.s16 q0, q0, q1
+; 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 <4 x i32> @llvm.arm.mve.mull.int.predicated.v4i32.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, i32 0, <8 x i1> %1, <4 x i32> undef)
+  ret <4 x i32> %2
+}
+
+define arm_aapcs_vfpcc <2 x i64> @test_vmullbq_int_x_u32(<4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vmullbq_int_x_u32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmullbt.s32 q2, q0, q1
+; CHECK-NEXT:    vmov q0, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <2 x i64> @llvm.arm.mve.mull.int.predicated.v2i64.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, i32 0, <4 x i1> %1, <2 x i64> undef)
+  ret <2 x i64> %2
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vmullbq_poly_x_p16(<8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vmullbq_poly_x_p16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmullbt.p16 q0, q0, q1
+; 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 <4 x i32> @llvm.arm.mve.mull.poly.predicated.v4i32.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, i32 0, <8 x i1> %1, <4 x i32> undef)
+  ret <4 x i32> %2
+}
+
+declare <4 x i32> @llvm.arm.mve.mull.poly.predicated.v4i32.v8i16.v8i1(<8 x i16>, <8 x i16>, i32, <8 x i1>, <4 x i32>) #1
+
diff --git a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmulltq.ll b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmulltq.ll
index 5dde90e298a..ca2aae6034a 100644
--- a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmulltq.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmulltq.ll
@@ -50,7 +50,7 @@ entry:
 
 declare <4 x i32> @llvm.arm.mve.vmull.poly.v4i32.v8i16(<8 x i16>, <8 x i16>, i32) #1
 
-define arm_aapcs_vfpcc <8 x i16> @test_vmulltq_int_m_s8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #0 {
+define arm_aapcs_vfpcc <8 x i16> @test_vmulltq_int_m_s8(<8 x i16> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vmulltq_int_m_s8:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
@@ -60,15 +60,15 @@ define arm_aapcs_vfpcc <8 x i16> @test_vmulltq_int_m_s8(<16 x i8> %inactive, <16
 entry:
   %0 = zext i16 %p to i32
   %1 = tail call <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32 %0)
-  %2 = tail call <8 x i16> @llvm.arm.mve.mull.int.predicated.v8i16.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, i32 1, <16 x i1> %1, <16 x i8> %inactive)
+  %2 = tail call <8 x i16> @llvm.arm.mve.mull.int.predicated.v8i16.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, i32 1, <16 x i1> %1, <8 x i16> %inactive)
   ret <8 x i16> %2
 }
 
 declare <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32) #1
 
-declare <8 x i16> @llvm.arm.mve.mull.int.predicated.v8i16.v16i8.v16i1(<16 x i8>, <16 x i8>, i32, <16 x i1>, <16 x i8>) #1
+declare <8 x i16> @llvm.arm.mve.mull.int.predicated.v8i16.v16i8.v16i1(<16 x i8>, <16 x i8>, i32, <16 x i1>, <8 x i16>) #1
 
-define arm_aapcs_vfpcc <4 x i32> @test_vmulltq_int_m_u16(<8 x i16> %inactive, <8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #0 {
+define arm_aapcs_vfpcc <4 x i32> @test_vmulltq_int_m_u16(<4 x i32> %inactive, <8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vmulltq_int_m_u16:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
@@ -78,15 +78,15 @@ define arm_aapcs_vfpcc <4 x i32> @test_vmulltq_int_m_u16(<8 x i16> %inactive, <8
 entry:
   %0 = zext i16 %p to i32
   %1 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %0)
-  %2 = tail call <4 x i32> @llvm.arm.mve.mull.int.predicated.v4i32.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, i32 1, <8 x i1> %1, <8 x i16> %inactive)
+  %2 = tail call <4 x i32> @llvm.arm.mve.mull.int.predicated.v4i32.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, i32 1, <8 x i1> %1, <4 x i32> %inactive)
   ret <4 x i32> %2
 }
 
 declare <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32) #1
 
-declare <4 x i32> @llvm.arm.mve.mull.int.predicated.v4i32.v8i16.v8i1(<8 x i16>, <8 x i16>, i32, <8 x i1>, <8 x i16>) #1
+declare <4 x i32> @llvm.arm.mve.mull.int.predicated.v4i32.v8i16.v8i1(<8 x i16>, <8 x i16>, i32, <8 x i1>, <4 x i32>) #1
 
-define arm_aapcs_vfpcc <2 x i64> @test_vmulltq_int_m_s32(<4 x i32> %inactive, <4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #0 {
+define arm_aapcs_vfpcc <2 x i64> @test_vmulltq_int_m_s32(<2 x i64> %inactive, <4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vmulltq_int_m_s32:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
@@ -96,15 +96,15 @@ define arm_aapcs_vfpcc <2 x i64> @test_vmulltq_int_m_s32(<4 x i32> %inactive, <4
 entry:
   %0 = zext i16 %p to i32
   %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
-  %2 = tail call <2 x i64> @llvm.arm.mve.mull.int.predicated.v2i64.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, i32 1, <4 x i1> %1, <4 x i32> %inactive)
+  %2 = tail call <2 x i64> @llvm.arm.mve.mull.int.predicated.v2i64.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, i32 1, <4 x i1> %1, <2 x i64> %inactive)
   ret <2 x i64> %2
 }
 
 declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32) #1
 
-declare <2 x i64> @llvm.arm.mve.mull.int.predicated.v2i64.v4i32.v4i1(<4 x i32>, <4 x i32>, i32, <4 x i1>, <4 x i32>) #1
+declare <2 x i64> @llvm.arm.mve.mull.int.predicated.v2i64.v4i32.v4i1(<4 x i32>, <4 x i32>, i32, <4 x i1>, <2 x i64>) #1
 
-define arm_aapcs_vfpcc <8 x i16> @test_vmulltq_poly_m_p8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #0 {
+define arm_aapcs_vfpcc <8 x i16> @test_vmulltq_poly_m_p8(<8 x i16> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vmulltq_poly_m_p8:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
@@ -114,8 +114,66 @@ define arm_aapcs_vfpcc <8 x i16> @test_vmulltq_poly_m_p8(<16 x i8> %inactive, <1
 entry:
   %0 = zext i16 %p to i32
   %1 = tail call <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32 %0)
-  %2 = tail call <8 x i16> @llvm.arm.mve.mull.poly.predicated.v8i16.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, i32 1, <16 x i1> %1, <16 x i8> %inactive)
+  %2 = tail call <8 x i16> @llvm.arm.mve.mull.poly.predicated.v8i16.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, i32 1, <16 x i1> %1, <8 x i16> %inactive)
+  ret <8 x i16> %2
+}
+
+declare <8 x i16> @llvm.arm.mve.mull.poly.predicated.v8i16.v16i8.v16i1(<16 x i8>, <16 x i8>, i32, <16 x i1>, <8 x i16>) #1
+
+define arm_aapcs_vfpcc <8 x i16> @test_vmulltq_int_x_u8(<16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vmulltq_int_x_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmulltt.s8 q0, q0, q1
+; 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 <8 x i16> @llvm.arm.mve.mull.int.predicated.v8i16.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, i32 1, <16 x i1> %1, <8 x i16> undef)
+  ret <8 x i16> %2
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vmulltq_int_x_s16(<8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vmulltq_int_x_s16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmulltt.s16 q0, q0, q1
+; 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 <4 x i32> @llvm.arm.mve.mull.int.predicated.v4i32.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, i32 1, <8 x i1> %1, <4 x i32> undef)
+  ret <4 x i32> %2
+}
+
+define arm_aapcs_vfpcc <2 x i64> @test_vmulltq_int_x_u32(<4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vmulltq_int_x_u32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmulltt.s32 q2, q0, q1
+; CHECK-NEXT:    vmov q0, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <2 x i64> @llvm.arm.mve.mull.int.predicated.v2i64.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, i32 1, <4 x i1> %1, <2 x i64> undef)
+  ret <2 x i64> %2
+}
+
+define arm_aapcs_vfpcc <8 x i16> @test_vmulltq_poly_x_p8(<16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vmulltq_poly_x_p8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmulltt.p8 q0, q0, q1
+; 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 <8 x i16> @llvm.arm.mve.mull.poly.predicated.v8i16.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, i32 1, <16 x i1> %1, <8 x i16> undef)
   ret <8 x i16> %2
 }
 
-declare <8 x i16> @llvm.arm.mve.mull.poly.predicated.v8i16.v16i8.v16i1(<16 x i8>, <16 x i8>, i32, <16 x i1>, <16 x i8>) #1
diff --git a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmulq.ll b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmulq.ll
index 09d8e11a71a..19ed08a77a2 100644
--- a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmulq.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vmulq.ll
@@ -1,7 +1,27 @@
 ; 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 <4 x i32> @test_vmulq_u32(<4 x i32> %a, <4 x i32> %b) {
+define arm_aapcs_vfpcc <16 x i8> @test_vmulq_u8(<16 x i8> %a, <16 x i8> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vmulq_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmul.i8 q0, q1, q0
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = mul <16 x i8> %b, %a
+  ret <16 x i8> %0
+}
+
+define arm_aapcs_vfpcc <8 x i16> @test_vmulq_s16(<8 x i16> %a, <8 x i16> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vmulq_s16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmul.i16 q0, q1, q0
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = mul <8 x i16> %b, %a
+  ret <8 x i16> %0
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vmulq_u32(<4 x i32> %a, <4 x i32> %b) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vmulq_u32:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmul.i32 q0, q1, q0
@@ -11,17 +31,17 @@ entry:
   ret <4 x i32> %0
 }
 
-define arm_aapcs_vfpcc <4 x float> @test_vmulq_f32(<4 x float> %a, <4 x float> %b) {
+define arm_aapcs_vfpcc <4 x float> @test_vmulq_f32(<4 x float> %a, <4 x float> %b) local_unnamed_addr #0 {
 ; CHECK-LABEL: test_vmulq_f32:
 ; CHECK:       @ %bb.0: @ %entry
-; CHECK-NEXT:    vmul.f32 q0, q1, q0
+; CHECK-NEXT:    vmul.f32 q0, q0, q1
 ; CHECK-NEXT:    bx lr
 entry:
-  %0 = fmul <4 x float> %b, %a
+  %0 = fmul <4 x float> %a, %b
   ret <4 x float> %0
 }
 
-define arm_aapcs_vfpcc <16 x i8> @test_vmulq_m_s8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) {
+define arm_aapcs_vfpcc <16 x i8> @test_vmulq_m_s8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
 ; CHECK-LABEL: test_vmulq_m_s8:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
@@ -35,11 +55,47 @@ entry:
   ret <16 x i8> %2
 }
 
-declare <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32)
+declare <16 x i1> @llvm.arm.mve.pred.i2v.v16i1(i32) #2
+
+declare <16 x i8> @llvm.arm.mve.mul.predicated.v16i8.v16i1(<16 x i8>, <16 x i8>, <16 x i1>, <16 x i8>) #2
+
+define arm_aapcs_vfpcc <8 x i16> @test_vmulq_m_u16(<8 x i16> %inactive, <8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vmulq_m_u16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmult.i16 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.mul.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 <16 x i8> @llvm.arm.mve.mul.predicated.v16i8.v16i1(<16 x i8>, <16 x i8>, <16 x i1>, <16 x i8>)
+declare <8 x i16> @llvm.arm.mve.mul.predicated.v8i16.v8i1(<8 x i16>, <8 x i16>, <8 x i1>, <8 x i16>) #2
 
-define arm_aapcs_vfpcc <8 x half> @test_vmulq_m_f16(<8 x half> %inactive, <8 x half> %a, <8 x half> %b, i16 zeroext %p) {
+define arm_aapcs_vfpcc <4 x i32> @test_vmulq_m_s32(<4 x i32> %inactive, <4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vmulq_m_s32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmult.i32 q0, q1, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x i32> @llvm.arm.mve.mul.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> %inactive)
+  ret <4 x i32> %2
+}
+
+declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32) #2
+
+declare <4 x i32> @llvm.arm.mve.mul.predicated.v4i32.v4i1(<4 x i32>, <4 x i32>, <4 x i1>, <4 x i32>) #2
+
+define arm_aapcs_vfpcc <8 x half> @test_vmulq_m_f16(<8 x half> %inactive, <8 x half> %a, <8 x half> %b, i16 zeroext %p) local_unnamed_addr #1 {
 ; CHECK-LABEL: test_vmulq_m_f16:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    vmsr p0, r0
@@ -53,6 +109,63 @@ entry:
   ret <8 x half> %2
 }
 
-declare <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32)
+declare <8 x half> @llvm.arm.mve.mul.predicated.v8f16.v8i1(<8 x half>, <8 x half>, <8 x i1>, <8 x half>) #2
+
+define arm_aapcs_vfpcc <16 x i8> @test_vmulq_x_u8(<16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vmulq_x_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmult.i8 q0, q0, q1
+; 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.mul.predicated.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, <16 x i1> %1, <16 x i8> undef)
+  ret <16 x i8> %2
+}
+
+define arm_aapcs_vfpcc <8 x i16> @test_vmulq_x_s16(<8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vmulq_x_s16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmult.i16 q0, q0, q1
+; 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.mul.predicated.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, <8 x i1> %1, <8 x i16> undef)
+  ret <8 x i16> %2
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vmulq_x_u32(<4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vmulq_x_u32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmult.i32 q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x i32> @llvm.arm.mve.mul.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> undef)
+  ret <4 x i32> %2
+}
+
+define arm_aapcs_vfpcc <4 x float> @test_vmulq_m_f32(<4 x float> %a, <4 x float> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vmulq_m_f32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vmult.f32 q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x float> @llvm.arm.mve.mul.predicated.v4f32.v4i1(<4 x float> %a, <4 x float> %b, <4 x i1> %1, <4 x float> undef)
+  ret <4 x float> %2
+}
+
+declare <4 x float> @llvm.arm.mve.mul.predicated.v4f32.v4i1(<4 x float>, <4 x float>, <4 x i1>, <4 x float>) #2
 
-declare <8 x half> @llvm.arm.mve.mul.predicated.v8f16.v8i1(<8 x half>, <8 x half>, <8 x i1>, <8 x half>)
diff --git a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vornq.ll b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vornq.ll
index 48f6a3cd23a..be4ad54dd13 100644
--- a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vornq.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vornq.ll
@@ -12,26 +12,26 @@ entry:
   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:
+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 <4 x i32> %b, <i32 -1, i32 -1, i32 -1, i32 -1>
-  %1 = or <4 x i32> %0, %a
-  ret <4 x i32> %1
+  %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 <8 x i16> @test_vornq_s16(<8 x i16> %a, <8 x i16> %b) local_unnamed_addr #0 {
-; CHECK-LABEL: test_vornq_s16:
+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 <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
+  %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 <4 x float> @test_vornq_f32(<4 x float> %a, <4 x float> %b) local_unnamed_addr #0 {
@@ -84,25 +84,98 @@ 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:
+define arm_aapcs_vfpcc <4 x i32> @test_vornq_m_s32(<4 x i32> %inactive, <4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vornq_m_s32:
 ; 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 <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x i32> @llvm.arm.mve.orn.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> %inactive)
+  ret <4 x i32> %2
+}
+
+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
+
+define arm_aapcs_vfpcc <8 x half> @test_vornq_m_f16(<8 x half> %inactive, <8 x half> %a, <8 x half> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vornq_m_f16:
+; 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 <8 x half> %a to <8 x i16>
+  %1 = bitcast <8 x half> %b to <8 x i16>
+  %2 = zext i16 %p to i32
+  %3 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %2)
+  %4 = bitcast <8 x half> %inactive to <8 x i16>
+  %5 = tail call <8 x i16> @llvm.arm.mve.orn.predicated.v8i16.v8i1(<8 x i16> %0, <8 x i16> %1, <8 x i1> %3, <8 x i16> %4)
+  %6 = bitcast <8 x i16> %5 to <8 x half>
+  ret <8 x half> %6
+}
+
+define arm_aapcs_vfpcc <16 x i8> @test_vornq_x_u8(<16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vornq_x_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vornt q0, q0, q1
+; 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> undef)
+  ret <16 x i8> %2
+}
+
+define arm_aapcs_vfpcc <8 x i16> @test_vornq_x_s16(<8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vornq_x_s16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vornt q0, q0, q1
+; 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> undef)
+  ret <8 x i16> %2
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vornq_x_u32(<4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vornq_x_u32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vornt q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x i32> @llvm.arm.mve.orn.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> undef)
+  ret <4 x i32> %2
+}
+
+define arm_aapcs_vfpcc <4 x float> @test_vornq_m_f32(<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, 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 = 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
+  %4 = 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> undef)
+  %5 = bitcast <4 x i32> %4 to <4 x float>
+  ret <4 x float> %5
 }
 
-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
diff --git a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vorrq.ll b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vorrq.ll
index ccb511a85e5..168dde0806b 100644
--- a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vorrq.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vorrq.ll
@@ -11,24 +11,24 @@ entry:
   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:
+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 <4 x i32> %b, %a
-  ret <4 x i32> %0
+  %0 = or <8 x i16> %b, %a
+  ret <8 x i16> %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:
+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 <8 x i16> %b, %a
-  ret <8 x i16> %0
+  %0 = or <4 x i32> %b, %a
+  ret <4 x i32> %0
 }
 
 define arm_aapcs_vfpcc <4 x float> @test_vorrq_f32(<4 x float> %a, <4 x float> %b) local_unnamed_addr #0 {
@@ -80,25 +80,98 @@ 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:
+define arm_aapcs_vfpcc <4 x i32> @test_vorrq_m_s32(<4 x i32> %inactive, <4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vorrq_m_s32:
+; 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 <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x i32> @llvm.arm.mve.orr.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> %inactive)
+  ret <4 x i32> %2
+}
+
+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
+
+define arm_aapcs_vfpcc <8 x half> @test_vorrq_m_f16(<8 x half> %inactive, <8 x half> %a, <8 x half> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vorrq_m_f16:
 ; 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 <8 x half> %a to <8 x i16>
+  %1 = bitcast <8 x half> %b to <8 x i16>
+  %2 = zext i16 %p to i32
+  %3 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %2)
+  %4 = bitcast <8 x half> %inactive to <8 x i16>
+  %5 = tail call <8 x i16> @llvm.arm.mve.orr.predicated.v8i16.v8i1(<8 x i16> %0, <8 x i16> %1, <8 x i1> %3, <8 x i16> %4)
+  %6 = bitcast <8 x i16> %5 to <8 x half>
+  ret <8 x half> %6
+}
+
+define arm_aapcs_vfpcc <16 x i8> @test_vorrq_x_u8(<16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vorrq_x_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vorrt q0, q0, q1
+; 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> undef)
+  ret <16 x i8> %2
+}
+
+define arm_aapcs_vfpcc <8 x i16> @test_vorrq_x_s16(<8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vorrq_x_s16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vorrt q0, q0, q1
+; 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> undef)
+  ret <8 x i16> %2
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vorrq_x_u32(<4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vorrq_x_u32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vorrt q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x i32> @llvm.arm.mve.orr.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> undef)
+  ret <4 x i32> %2
+}
+
+define arm_aapcs_vfpcc <4 x float> @test_vorrq_m_f32(<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, 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 = 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
+  %4 = 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> undef)
+  %5 = bitcast <4 x i32> %4 to <4 x float>
+  ret <4 x float> %5
 }
 
-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
diff --git a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vrhaddq.ll b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vrhaddq.ll
index 0ba47befa37..15b23d5ed27 100644
--- a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vrhaddq.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vrhaddq.ll
@@ -90,3 +90,46 @@ entry:
 declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32) #1
 
 declare <4 x i32> @llvm.arm.mve.rhadd.predicated.v4i32.v4i1(<4 x i32>, <4 x i32>, <4 x i1>, <4 x i32>) #1
+
+define arm_aapcs_vfpcc <16 x i8> @test_vrhaddq_x_u8(<16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vrhaddq_x_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vrhaddt.s8 q0, q0, q1
+; 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.rhadd.predicated.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, <16 x i1> %1, <16 x i8> undef)
+  ret <16 x i8> %2
+}
+
+define arm_aapcs_vfpcc <8 x i16> @test_vrhaddq_x_u16(<8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vrhaddq_x_u16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vrhaddt.s16 q0, q0, q1
+; 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.rhadd.predicated.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, <8 x i1> %1, <8 x i16> undef)
+  ret <8 x i16> %2
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vrhaddq_x_u32(<4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vrhaddq_x_u32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vrhaddt.s32 q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x i32> @llvm.arm.mve.rhadd.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> undef)
+  ret <4 x i32> %2
+}
+
diff --git a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vrmulhq.ll b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vrmulhq.ll
index 3975e4eca87..9418f581eb4 100644
--- a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vrmulhq.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vrmulhq.ll
@@ -90,3 +90,46 @@ entry:
 declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32) #1
 
 declare <4 x i32> @llvm.arm.mve.rmulh.predicated.v4i32.v4i1(<4 x i32>, <4 x i32>, <4 x i1>, <4 x i32>) #1
+
+define arm_aapcs_vfpcc <16 x i8> @test_vrmulhq_x_u8(<16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vrmulhq_x_u8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vrmulht.s8 q0, q0, q1
+; 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.rmulh.predicated.v16i8.v16i1(<16 x i8> %a, <16 x i8> %b, <16 x i1> %1, <16 x i8> undef)
+  ret <16 x i8> %2
+}
+
+define arm_aapcs_vfpcc <8 x i16> @test_vrmulhq_x_s16(<8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vrmulhq_x_s16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vrmulht.s16 q0, q0, q1
+; 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.rmulh.predicated.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, <8 x i1> %1, <8 x i16> undef)
+  ret <8 x i16> %2
+}
+
+define arm_aapcs_vfpcc <4 x i32> @test_vrmulhq_m_u32(<4 x i32> %a, <4 x i32> %b, i16 zeroext %p) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vrmulhq_m_u32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vrmulht.s32 q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x i32> @llvm.arm.mve.rmulh.predicated.v4i32.v4i1(<4 x i32> %a, <4 x i32> %b, <4 x i1> %1, <4 x i32> undef)
+  ret <4 x i32> %2
+}
+
diff --git a/llvm/test/CodeGen/Thumb2/mve-intrinsics/vsubq.ll b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vsubq.ll
new file mode 100644
index 00000000000..2959e61e0d8
--- /dev/null
+++ b/llvm/test/CodeGen/Thumb2/mve-intrinsics/vsubq.ll
@@ -0,0 +1,93 @@
+; 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 <4 x i32> @test_vsubq_u32(<4 x i32> %a, <4 x i32> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vsubq_u32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vsub.i32 q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = sub <4 x i32> %a, %b
+  ret <4 x i32> %0
+}
+
+define arm_aapcs_vfpcc <8 x half> @test_vsubq_f16(<8 x half> %a, <8 x half> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_vsubq_f16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vsub.f16 q0, q0, q1
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = fsub <8 x half> %a, %b
+  ret <8 x half> %0
+}
+
+define arm_aapcs_vfpcc <16 x i8> @test_vsubq_m_s8(<16 x i8> %inactive, <16 x i8> %a, <16 x i8> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vsubq_m_s8:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vsubt.i8 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.sub.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.sub.predicated.v16i8.v16i1(<16 x i8>, <16 x i8>, <16 x i1>, <16 x i8>) #2
+
+define arm_aapcs_vfpcc <4 x float> @test_vsubq_m_f32(<4 x float> %inactive, <4 x float> %a, <4 x float> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vsubq_m_f32:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vsubt.f32 q0, q1, q2
+; CHECK-NEXT:    bx lr
+entry:
+  %0 = zext i16 %p to i32
+  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
+  %2 = tail call <4 x float> @llvm.arm.mve.sub.predicated.v4f32.v4i1(<4 x float> %a, <4 x float> %b, <4 x i1> %1, <4 x float> %inactive)
+  ret <4 x float> %2
+}
+
+declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32) #2
+
+declare <4 x float> @llvm.arm.mve.sub.predicated.v4f32.v4i1(<4 x float>, <4 x float>, <4 x i1>, <4 x float>) #2
+
+define arm_aapcs_vfpcc <8 x i16> @test_vsubq_x_u16(<8 x i16> %a, <8 x i16> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vsubq_x_u16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vsubt.i16 q0, q0, q1
+; 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.sub.predicated.v8i16.v8i1(<8 x i16> %a, <8 x i16> %b, <8 x i1> %1, <8 x i16> undef)
+  ret <8 x i16> %2
+}
+
+declare <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32) #2
+
+declare <8 x i16> @llvm.arm.mve.sub.predicated.v8i16.v8i1(<8 x i16>, <8 x i16>, <8 x i1>, <8 x i16>) #2
+
+define arm_aapcs_vfpcc <8 x half> @test_vsubq_x_f16(<8 x half> %a, <8 x half> %b, i16 zeroext %p) local_unnamed_addr #1 {
+; CHECK-LABEL: test_vsubq_x_f16:
+; CHECK:       @ %bb.0: @ %entry
+; CHECK-NEXT:    vmsr p0, r0
+; CHECK-NEXT:    vpst
+; CHECK-NEXT:    vsubt.f16 q0, q0, q1
+; 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 half> @llvm.arm.mve.sub.predicated.v8f16.v8i1(<8 x half> %a, <8 x half> %b, <8 x i1> %1, <8 x half> undef)
+  ret <8 x half> %2
+}
+
+declare <8 x half> @llvm.arm.mve.sub.predicated.v8f16.v8i1(<8 x half>, <8 x half>, <8 x i1>, <8 x half>) #2
+