[ARM/AArch64] Relocate and update InterleavedAccessPass tests (NFC)

The interleaved access pass is an IR-to-IR transformation that runs before code
generation. It matches interleaved memory operations to target-specific
intrinsics (that are later lowered to load and store multiple instructions on
ARM/AArch64). We place tests for similar passes (e.g., GlobalMergePass) under
test/Transforms. This patch moves the InterleavedAccessPass tests out of
test/CodeGen and into target-specific directories under
test/Transforms/InterleavedAccess.

Although the pass is an IR pass, many of the existing tests were llc tests
rather opt tests. For example, the tests would check for ldN/stN instructions
generated by llc rather than the intrinsic calls the pass actually inserts.
Thus, this patch updates all tests to be opt tests that check for the inserted
intrinsics. We already have separate CodeGen tests that ensure we lower the
interleaved access intrinsics to their corresponding ldN/stN instructions. In
addition to migrating the tests to opt, this patch also performs some minor
clean-up (to ensure consistent naming, etc.).

Differential Revision: https://reviews.llvm.org/D29184

llvm-svn: 293309

6 files changed, 1407 insertions, 0 deletions

diff --git a/llvm/test/Transforms/InterleavedAccess/AArch64/interleaved-accesses-extract-user.ll b/llvm/test/Transforms/InterleavedAccess/AArch64/interleaved-accesses-extract-user.ll
new file mode 100644
index 00000000000..744a36be80e
--- /dev/null
+++ b/llvm/test/Transforms/InterleavedAccess/AArch64/interleaved-accesses-extract-user.ll
@@ -0,0 +1,113 @@
+; RUN: opt < %s -interleaved-access -S | FileCheck %s
+
+target datalayout = "e-m:e-i64:64-i128:128-n32:64-S128"
+target triple = "aarch64--linux-gnu"
+
+define void @extract_user_basic(<8 x i32>* %ptr, i1 %c) {
+; CHECK-LABEL: @extract_user_basic(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <8 x i32>* %ptr to <4 x i32>*
+; CHECK-NEXT:    [[LDN:%.*]] = call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2.v4i32.p0v4i32(<4 x i32>* [[TMP0]])
+; CHECK-NEXT:    [[TMP1:%.*]] = extractvalue { <4 x i32>, <4 x i32> } [[LDN]], 0
+; CHECK-NEXT:    br i1 %c, label %if.then, label %if.merge
+; CHECK:       if.then:
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <4 x i32> [[TMP1]], i64 1
+; CHECK-NEXT:    br label %if.merge
+; CHECK:       if.merge:
+; CHECK-NEXT:    ret void
+;
+entry:
+  %interleaved.vec = load <8 x i32>, <8 x i32>* %ptr, align 8
+  %v0 = shufflevector <8 x i32> %interleaved.vec, <8 x i32> undef, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
+  br i1 %c, label %if.then, label %if.merge
+
+if.then:
+  %e0 = extractelement <8 x i32> %interleaved.vec, i32 2
+  br label %if.merge
+
+if.merge:
+  ret void
+}
+
+define void @extract_user_multi(<8 x i32>* %ptr, i1 %c) {
+; CHECK-LABEL: @extract_user_multi(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <8 x i32>* %ptr to <4 x i32>*
+; CHECK-NEXT:    [[LDN:%.*]] = call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2.v4i32.p0v4i32(<4 x i32>* [[TMP0]])
+; CHECK-NEXT:    [[TMP1:%.*]] = extractvalue { <4 x i32>, <4 x i32> } [[LDN]], 0
+; CHECK-NEXT:    br i1 %c, label %if.then, label %if.merge
+; CHECK:       if.then:
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <4 x i32> [[TMP1]], i64 0
+; CHECK-NEXT:    br label %if.merge
+; CHECK:       if.merge:
+; CHECK-NEXT:    [[TMP3:%.*]] = extractelement <4 x i32> [[TMP1]], i64 1
+; CHECK-NEXT:    ret void
+;
+entry:
+  %interleaved.vec = load <8 x i32>, <8 x i32>* %ptr, align 8
+  %v0 = shufflevector <8 x i32> %interleaved.vec, <8 x i32> undef, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
+  br i1 %c, label %if.then, label %if.merge
+
+if.then:
+  %e0 = extractelement <8 x i32> %interleaved.vec, i32 0
+  br label %if.merge
+
+if.merge:
+  %e1 = extractelement <8 x i32> %interleaved.vec, i32 2
+  ret void
+}
+
+define void @extract_user_multi_no_dom(<8 x i32>* %ptr, i1 %c) {
+; CHECK-LABEL: @extract_user_multi_no_dom(
+; CHECK-NOT:     @llvm.aarch64.neon
+; CHECK:         ret void
+;
+entry:
+  %interleaved.vec = load <8 x i32>, <8 x i32>* %ptr, align 8
+  %e0 = extractelement <8 x i32> %interleaved.vec, i32 0
+  br i1 %c, label %if.then, label %if.merge
+
+if.then:
+  %v0 = shufflevector <8 x i32> %interleaved.vec, <8 x i32> undef, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
+  %e1 = extractelement <8 x i32> %interleaved.vec, i32 2
+  br label %if.merge
+
+if.merge:
+  ret void
+}
+
+define void @extract_user_wrong_const_index(<8 x i32>* %ptr) {
+; CHECK-LABEL: @extract_user_wrong_const_index(
+; CHECK-NOT:     @llvm.aarch64.neon
+; CHECK:         ret void
+;
+entry:
+  %interleaved.vec = load <8 x i32>, <8 x i32>* %ptr, align 8
+  %v0 = shufflevector <8 x i32> %interleaved.vec, <8 x i32> undef, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
+  %e0 = extractelement <8 x i32> %interleaved.vec, i32 1
+  ret void
+}
+
+define void @extract_user_undef_index(<8 x i32>* %ptr) {
+; CHECK-LABEL: @extract_user_undef_index(
+; CHECK-NOT:     @llvm.aarch64.neon
+; CHECK:         ret void
+;
+entry:
+  %interleaved.vec = load <8 x i32>, <8 x i32>* %ptr, align 8
+  %v0 = shufflevector <8 x i32> %interleaved.vec, <8 x i32> undef, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
+  %e0 = extractelement <8 x i32> %interleaved.vec, i32 undef
+  ret void
+}
+
+define void @extract_user_var_index(<8 x i32>* %ptr, i32 %i) {
+; CHECK-LABEL: @extract_user_var_index(
+; CHECK-NOT:     @llvm.aarch64.neon
+; CHECK:         ret void
+;
+entry:
+  %interleaved.vec = load <8 x i32>, <8 x i32>* %ptr, align 8
+  %v0 = shufflevector <8 x i32> %interleaved.vec, <8 x i32> undef, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
+  %e0 = extractelement <8 x i32> %interleaved.vec, i32 %i
+  ret void
+}
diff --git a/llvm/test/Transforms/InterleavedAccess/AArch64/interleaved-accesses.ll b/llvm/test/Transforms/InterleavedAccess/AArch64/interleaved-accesses.ll
new file mode 100644
index 00000000000..702dfdbb81a
--- /dev/null
+++ b/llvm/test/Transforms/InterleavedAccess/AArch64/interleaved-accesses.ll
@@ -0,0 +1,549 @@
+; RUN: opt < %s -interleaved-access -S | FileCheck %s -check-prefix=NEON
+; RUN: opt < %s -mattr=-neon -interleaved-access -S | FileCheck %s -check-prefix=NO_NEON
+
+target datalayout = "e-m:e-i64:64-i128:128-n32:64-S128"
+target triple = "aarch64--linux-gnu"
+
+define void @load_factor2(<16 x i8>* %ptr) {
+; NEON-LABEL:    @load_factor2(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <16 x i8>* %ptr to <8 x i8>*
+; NEON-NEXT:       [[LDN:%.*]] = call { <8 x i8>, <8 x i8> } @llvm.aarch64.neon.ld2.v8i8.p0v8i8(<8 x i8>* [[TMP1]])
+; NEON-NEXT:       [[TMP2:%.*]] = extractvalue { <8 x i8>, <8 x i8> } [[LDN]], 1
+; NEON-NEXT:       [[TMP3:%.*]] = extractvalue { <8 x i8>, <8 x i8> } [[LDN]], 0
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @load_factor2(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <16 x i8>, <16 x i8>* %ptr, align 4
+  %v0 = shufflevector <16 x i8> %interleaved.vec, <16 x i8> undef, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14>
+  %v1 = shufflevector <16 x i8> %interleaved.vec, <16 x i8> undef, <8 x i32> <i32 1, i32 3, i32 5, i32 7, i32 9, i32 11, i32 13, i32 15>
+  ret void
+}
+
+define void @load_factor3(<12 x i32>* %ptr) {
+; NEON-LABEL:    @load_factor3(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <12 x i32>* %ptr to <4 x i32>*
+; NEON-NEXT:       [[LDN:%.*]] = call { <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld3.v4i32.p0v4i32(<4 x i32>* [[TMP1]])
+; NEON-NEXT:       [[TMP2:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32> } [[LDN]], 2
+; NEON-NEXT:       [[TMP3:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32> } [[LDN]], 1
+; NEON-NEXT:       [[TMP4:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32> } [[LDN]], 0
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @load_factor3(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <12 x i32>, <12 x i32>* %ptr, align 4
+  %v0 = shufflevector <12 x i32> %interleaved.vec, <12 x i32> undef, <4 x i32> <i32 0, i32 3, i32 6, i32 9>
+  %v1 = shufflevector <12 x i32> %interleaved.vec, <12 x i32> undef, <4 x i32> <i32 1, i32 4, i32 7, i32 10>
+  %v2 = shufflevector <12 x i32> %interleaved.vec, <12 x i32> undef, <4 x i32> <i32 2, i32 5, i32 8, i32 11>
+  ret void
+}
+
+define void @load_factor4(<16 x i32>* %ptr) {
+; NEON-LABEL:    @load_factor4(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <16 x i32>* %ptr to <4 x i32>*
+; NEON-NEXT:       [[LDN:%.*]] = call { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld4.v4i32.p0v4i32(<4 x i32>* [[TMP1]])
+; NEON-NEXT:       [[TMP2:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[LDN]], 3
+; NEON-NEXT:       [[TMP3:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[LDN]], 2
+; NEON-NEXT:       [[TMP4:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[LDN]], 1
+; NEON-NEXT:       [[TMP5:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[LDN]], 0
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @load_factor4(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <16 x i32>, <16 x i32>* %ptr, align 4
+  %v0 = shufflevector <16 x i32> %interleaved.vec, <16 x i32> undef, <4 x i32> <i32 0, i32 4, i32 8, i32 12>
+  %v1 = shufflevector <16 x i32> %interleaved.vec, <16 x i32> undef, <4 x i32> <i32 1, i32 5, i32 9, i32 13>
+  %v2 = shufflevector <16 x i32> %interleaved.vec, <16 x i32> undef, <4 x i32> <i32 2, i32 6, i32 10, i32 14>
+  %v3 = shufflevector <16 x i32> %interleaved.vec, <16 x i32> undef, <4 x i32> <i32 3, i32 7, i32 11, i32 15>
+  ret void
+}
+
+define void @store_factor2(<16 x i8>* %ptr, <8 x i8> %v0, <8 x i8> %v1) {
+; NEON-LABEL:    @store_factor2(
+; NEON-NEXT:       [[TMP1:%.*]] = shufflevector <8 x i8> %v0, <8 x i8> %v1, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <8 x i8> %v0, <8 x i8> %v1, <8 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
+; NEON-NEXT:       [[TMP3:%.*]] = bitcast <16 x i8>* %ptr to <8 x i8>*
+; NEON-NEXT:       call void @llvm.aarch64.neon.st2.v8i8.p0v8i8(<8 x i8> [[TMP1]], <8 x i8> [[TMP2]], <8 x i8>* [[TMP3]])
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_factor2(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <8 x i8> %v0, <8 x i8> %v1, <16 x i32> <i32 0, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11, i32 4, i32 12, i32 5, i32 13, i32 6, i32 14, i32 7, i32 15>
+  store <16 x i8> %interleaved.vec, <16 x i8>* %ptr, align 4
+  ret void
+}
+
+define void @store_factor3(<12 x i32>* %ptr, <4 x i32> %v0, <4 x i32> %v1, <4 x i32> %v2) {
+; NEON-LABEL:    @store_factor3(
+; NEON:            [[TMP1:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 8, i32 9, i32 10, i32 11>
+; NEON-NEXT:       [[TMP4:%.*]] = bitcast <12 x i32>* %ptr to <4 x i32>*
+; NEON-NEXT:       call void @llvm.aarch64.neon.st3.v4i32.p0v4i32(<4 x i32> [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> [[TMP3]], <4 x i32>* [[TMP4]])
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_factor3(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %s0 = shufflevector <4 x i32> %v0, <4 x i32> %v1, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %s1 = shufflevector <4 x i32> %v2, <4 x i32> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef>
+  %interleaved.vec = shufflevector <8 x i32> %s0, <8 x i32> %s1, <12 x i32> <i32 0, i32 4, i32 8, i32 1, i32 5, i32 9, i32 2, i32 6, i32 10, i32 3, i32 7, i32 11>
+  store <12 x i32> %interleaved.vec, <12 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_factor4(<16 x i32>* %ptr, <4 x i32> %v0, <4 x i32> %v1, <4 x i32> %v2, <4 x i32> %v3) {
+; NEON-LABEL:    @store_factor4(
+; NEON:            [[TMP1:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 8, i32 9, i32 10, i32 11>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 12, i32 13, i32 14, i32 15>
+; NEON-NEXT:       [[TMP5:%.*]] = bitcast <16 x i32>* %ptr to <4 x i32>*
+; NEON-NEXT:       call void @llvm.aarch64.neon.st4.v4i32.p0v4i32(<4 x i32> [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> [[TMP3]], <4 x i32> [[TMP4]], <4 x i32>* [[TMP5]])
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_factor4(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %s0 = shufflevector <4 x i32> %v0, <4 x i32> %v1, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %s1 = shufflevector <4 x i32> %v2, <4 x i32> %v3, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %interleaved.vec = shufflevector <8 x i32> %s0, <8 x i32> %s1, <16 x i32> <i32 0, i32 4, i32 8, i32 12, i32 1, i32 5, i32 9, i32 13, i32 2, i32 6, i32 10, i32 14, i32 3, i32 7, i32 11, i32 15>
+  store <16 x i32> %interleaved.vec, <16 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @load_ptrvec_factor2(<4 x i32*>* %ptr) {
+; NEON-LABEL:    @load_ptrvec_factor2(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <4 x i32*>* %ptr to <2 x i64>*
+; NEON-NEXT:       [[LDN:%.*]] = call { <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld2.v2i64.p0v2i64(<2 x i64>* [[TMP1]])
+; NEON-NEXT:       [[TMP2:%.*]] = extractvalue { <2 x i64>, <2 x i64> } [[LDN]], 1
+; NEON-NEXT:       [[TMP3:%.*]] = inttoptr <2 x i64> [[TMP2]] to <2 x i32*>
+; NEON-NEXT:       [[TMP4:%.*]] = extractvalue { <2 x i64>, <2 x i64> } [[LDN]], 0
+; NEON-NEXT:       [[TMP5:%.*]] = inttoptr <2 x i64> [[TMP4]] to <2 x i32*>
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @load_ptrvec_factor2(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <4 x i32*>, <4 x i32*>* %ptr, align 4
+  %v0 = shufflevector <4 x i32*> %interleaved.vec, <4 x i32*> undef, <2 x i32> <i32 0, i32 2>
+  %v1 = shufflevector <4 x i32*> %interleaved.vec, <4 x i32*> undef, <2 x i32> <i32 1, i32 3>
+  ret void
+}
+
+define void @load_ptrvec_factor3(<6 x i32*>* %ptr) {
+; NEON-LABEL:    @load_ptrvec_factor3(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <6 x i32*>* %ptr to <2 x i64>*
+; NEON-NEXT:       [[LDN:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld3.v2i64.p0v2i64(<2 x i64>* [[TMP1]])
+; NEON-NEXT:       [[TMP2:%.*]] = extractvalue { <2 x i64>, <2 x i64>, <2 x i64> } [[LDN]], 2
+; NEON-NEXT:       [[TMP3:%.*]] = inttoptr <2 x i64> [[TMP2]] to <2 x i32*>
+; NEON-NEXT:       [[TMP4:%.*]] = extractvalue { <2 x i64>, <2 x i64>, <2 x i64> } [[LDN]], 1
+; NEON-NEXT:       [[TMP5:%.*]] = inttoptr <2 x i64> [[TMP4]] to <2 x i32*>
+; NEON-NEXT:       [[TMP6:%.*]] = extractvalue { <2 x i64>, <2 x i64>, <2 x i64> } [[LDN]], 0
+; NEON-NEXT:       [[TMP7:%.*]] = inttoptr <2 x i64> [[TMP6]] to <2 x i32*>
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @load_ptrvec_factor3(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <6 x i32*>, <6 x i32*>* %ptr, align 4
+  %v0 = shufflevector <6 x i32*> %interleaved.vec, <6 x i32*> undef, <2 x i32> <i32 0, i32 3>
+  %v1 = shufflevector <6 x i32*> %interleaved.vec, <6 x i32*> undef, <2 x i32> <i32 1, i32 4>
+  %v2 = shufflevector <6 x i32*> %interleaved.vec, <6 x i32*> undef, <2 x i32> <i32 2, i32 5>
+  ret void
+}
+
+define void @load_ptrvec_factor4(<8 x i32*>* %ptr) {
+; NEON-LABEL:    @load_ptrvec_factor4(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <8 x i32*>* %ptr to <2 x i64>*
+; NEON-NEXT:       [[LDN:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld4.v2i64.p0v2i64(<2 x i64>* [[TMP1]])
+; NEON-NEXT:       [[TMP2:%.*]] = extractvalue { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } [[LDN]], 3
+; NEON-NEXT:       [[TMP3:%.*]] = inttoptr <2 x i64> [[TMP2]] to <2 x i32*>
+; NEON-NEXT:       [[TMP4:%.*]] = extractvalue { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } [[LDN]], 2
+; NEON-NEXT:       [[TMP5:%.*]] = inttoptr <2 x i64> [[TMP4]] to <2 x i32*>
+; NEON-NEXT:       [[TMP6:%.*]] = extractvalue { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } [[LDN]], 1
+; NEON-NEXT:       [[TMP7:%.*]] = inttoptr <2 x i64> [[TMP6]] to <2 x i32*>
+; NEON-NEXT:       [[TMP8:%.*]] = extractvalue { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } [[LDN]], 0
+; NEON-NEXT:       [[TMP9:%.*]] = inttoptr <2 x i64> [[TMP8]] to <2 x i32*>
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @load_ptrvec_factor4(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <8 x i32*>, <8 x i32*>* %ptr, align 4
+  %v0 = shufflevector <8 x i32*> %interleaved.vec, <8 x i32*> undef, <2 x i32> <i32 0, i32 4>
+  %v1 = shufflevector <8 x i32*> %interleaved.vec, <8 x i32*> undef, <2 x i32> <i32 1, i32 5>
+  %v2 = shufflevector <8 x i32*> %interleaved.vec, <8 x i32*> undef, <2 x i32> <i32 2, i32 6>
+  %v3 = shufflevector <8 x i32*> %interleaved.vec, <8 x i32*> undef, <2 x i32> <i32 3, i32 7>
+  ret void
+}
+
+define void @store_ptrvec_factor2(<4 x i32*>* %ptr, <2 x i32*> %v0, <2 x i32*> %v1) {
+; NEON-LABEL:    @store_ptrvec_factor2(
+; NEON-NEXT:       [[TMP1:%.*]] = ptrtoint <2 x i32*> %v0 to <2 x i64>
+; NEON-NEXT:       [[TMP2:%.*]] = ptrtoint <2 x i32*> %v1 to <2 x i64>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <2 x i64> [[TMP1]], <2 x i64> [[TMP2]], <2 x i32> <i32 0, i32 1>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <2 x i64> [[TMP1]], <2 x i64> [[TMP2]], <2 x i32> <i32 2, i32 3>
+; NEON-NEXT:       [[TMP5:%.*]] = bitcast <4 x i32*>* %ptr to <2 x i64>*
+; NEON-NEXT:       call void @llvm.aarch64.neon.st2.v2i64.p0v2i64(<2 x i64> [[TMP3]], <2 x i64> [[TMP4]], <2 x i64>* [[TMP5]])
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_ptrvec_factor2(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <2 x i32*> %v0, <2 x i32*> %v1, <4 x i32> <i32 0, i32 2, i32 1, i32 3>
+  store <4 x i32*> %interleaved.vec, <4 x i32*>* %ptr, align 4
+  ret void
+}
+
+define void @store_ptrvec_factor3(<6 x i32*>* %ptr, <2 x i32*> %v0, <2 x i32*> %v1, <2 x i32*> %v2) {
+; NEON-LABEL:    @store_ptrvec_factor3(
+; NEON:            [[TMP1:%.*]] = ptrtoint <4 x i32*> %s0 to <4 x i64>
+; NEON-NEXT:       [[TMP2:%.*]] = ptrtoint <4 x i32*> %s1 to <4 x i64>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <4 x i64> [[TMP1]], <4 x i64> [[TMP2]], <2 x i32> <i32 0, i32 1>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <4 x i64> [[TMP1]], <4 x i64> [[TMP2]], <2 x i32> <i32 2, i32 3>
+; NEON-NEXT:       [[TMP5:%.*]] = shufflevector <4 x i64> [[TMP1]], <4 x i64> [[TMP2]], <2 x i32> <i32 4, i32 5>
+; NEON-NEXT:       [[TMP6:%.*]] = bitcast <6 x i32*>* %ptr to <2 x i64>*
+; NEON-NEXT:       call void @llvm.aarch64.neon.st3.v2i64.p0v2i64(<2 x i64> [[TMP3]], <2 x i64> [[TMP4]], <2 x i64> [[TMP5]], <2 x i64>* [[TMP6]])
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_ptrvec_factor3(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %s0 = shufflevector <2 x i32*> %v0, <2 x i32*> %v1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+  %s1 = shufflevector <2 x i32*> %v2, <2 x i32*> undef, <4 x i32> <i32 0, i32 1, i32 undef, i32 undef>
+  %interleaved.vec = shufflevector <4 x i32*> %s0, <4 x i32*> %s1, <6 x i32> <i32 0, i32 2, i32 4, i32 1, i32 3, i32 5>
+  store <6 x i32*> %interleaved.vec, <6 x i32*>* %ptr, align 4
+  ret void
+}
+
+define void @store_ptrvec_factor4(<8 x i32*>* %ptr, <2 x i32*> %v0, <2 x i32*> %v1, <2 x i32*> %v2, <2 x i32*> %v3) {
+; NEON-LABEL:    @store_ptrvec_factor4(
+; NEON:            [[TMP1:%.*]] = ptrtoint <4 x i32*> %s0 to <4 x i64>
+; NEON-NEXT:       [[TMP2:%.*]] = ptrtoint <4 x i32*> %s1 to <4 x i64>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <4 x i64> [[TMP1]], <4 x i64> [[TMP2]], <2 x i32> <i32 0, i32 1>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <4 x i64> [[TMP1]], <4 x i64> [[TMP2]], <2 x i32> <i32 2, i32 3>
+; NEON-NEXT:       [[TMP5:%.*]] = shufflevector <4 x i64> [[TMP1]], <4 x i64> [[TMP2]], <2 x i32> <i32 4, i32 5>
+; NEON-NEXT:       [[TMP6:%.*]] = shufflevector <4 x i64> [[TMP1]], <4 x i64> [[TMP2]], <2 x i32> <i32 6, i32 7>
+; NEON-NEXT:       [[TMP7:%.*]] = bitcast <8 x i32*>* %ptr to <2 x i64>*
+; NEON-NEXT:       call void @llvm.aarch64.neon.st4.v2i64.p0v2i64(<2 x i64> [[TMP3]], <2 x i64> [[TMP4]], <2 x i64> [[TMP5]], <2 x i64> [[TMP6]], <2 x i64>* [[TMP7]])
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_ptrvec_factor4(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %s0 = shufflevector <2 x i32*> %v0, <2 x i32*> %v1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+  %s1 = shufflevector <2 x i32*> %v2, <2 x i32*> %v3, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+  %interleaved.vec = shufflevector <4 x i32*> %s0, <4 x i32*> %s1, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 1, i32 3, i32 5, i32 7>
+  store <8 x i32*> %interleaved.vec, <8 x i32*>* %ptr, align 4
+  ret void
+}
+
+define void @load_undef_mask_factor2(<8 x i32>* %ptr) {
+; NEON-LABEL:    @load_undef_mask_factor2(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <8 x i32>* %ptr to <4 x i32>*
+; NEON-NEXT:       [[LDN:%.*]] = call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2.v4i32.p0v4i32(<4 x i32>* [[TMP1]])
+; NEON-NEXT:       [[TMP2:%.*]] = extractvalue { <4 x i32>, <4 x i32> } [[LDN]], 1
+; NEON-NEXT:       [[TMP3:%.*]] = extractvalue { <4 x i32>, <4 x i32> } [[LDN]], 0
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @load_undef_mask_factor2(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <8 x i32>, <8 x i32>* %ptr, align 4
+  %v0 = shufflevector <8 x i32> %interleaved.vec, <8 x i32> undef, <4 x i32> <i32 undef, i32 2, i32 undef, i32 6>
+  %v1 = shufflevector <8 x i32> %interleaved.vec, <8 x i32> undef, <4 x i32> <i32 undef, i32 3, i32 undef, i32 7>
+  ret void
+}
+
+define void @load_undef_mask_factor3(<12 x i32>* %ptr) {
+; NEON-LABEL:    @load_undef_mask_factor3(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <12 x i32>* %ptr to <4 x i32>*
+; NEON-NEXT:       [[LDN:%.*]] = call { <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld3.v4i32.p0v4i32(<4 x i32>* [[TMP1]])
+; NEON-NEXT:       [[TMP2:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32> } [[LDN]], 2
+; NEON-NEXT:       [[TMP3:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32> } [[LDN]], 1
+; NEON-NEXT:       [[TMP4:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32> } [[LDN]], 0
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @load_undef_mask_factor3(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <12 x i32>, <12 x i32>* %ptr, align 4
+  %v0 = shufflevector <12 x i32> %interleaved.vec, <12 x i32> undef, <4 x i32> <i32 0, i32 3, i32 6, i32 9>
+  %v1 = shufflevector <12 x i32> %interleaved.vec, <12 x i32> undef, <4 x i32> <i32 1, i32 4, i32 7, i32 10>
+  %v2 = shufflevector <12 x i32> %interleaved.vec, <12 x i32> undef, <4 x i32> <i32 2, i32 undef, i32 undef, i32 undef>
+  ret void
+}
+
+define void @load_undef_mask_factor4(<16 x i32>* %ptr) {
+; NEON-LABEL:    @load_undef_mask_factor4(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <16 x i32>* %ptr to <4 x i32>*
+; NEON-NEXT:       [[LDN:%.*]] = call { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld4.v4i32.p0v4i32(<4 x i32>* [[TMP1]])
+; NEON-NEXT:       [[TMP2:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[LDN]], 3
+; NEON-NEXT:       [[TMP3:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[LDN]], 2
+; NEON-NEXT:       [[TMP4:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[LDN]], 1
+; NEON-NEXT:       [[TMP5:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[LDN]], 0
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @load_undef_mask_factor4(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <16 x i32>, <16 x i32>* %ptr, align 4
+  %v0 = shufflevector <16 x i32> %interleaved.vec, <16 x i32> undef, <4 x i32> <i32 0, i32 4, i32 undef, i32 undef>
+  %v1 = shufflevector <16 x i32> %interleaved.vec, <16 x i32> undef, <4 x i32> <i32 1, i32 5, i32 undef, i32 undef>
+  %v2 = shufflevector <16 x i32> %interleaved.vec, <16 x i32> undef, <4 x i32> <i32 2, i32 6, i32 undef, i32 undef>
+  %v3 = shufflevector <16 x i32> %interleaved.vec, <16 x i32> undef, <4 x i32> <i32 3, i32 7, i32 undef, i32 undef>
+  ret void
+}
+
+define void @store_undef_mask_factor2(<8 x i32>* %ptr, <4 x i32> %v0, <4 x i32> %v1) {
+; NEON-LABEL:    @store_undef_mask_factor2(
+; NEON-NEXT:       [[TMP1:%.*]] = shufflevector <4 x i32> %v0, <4 x i32> %v1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <4 x i32> %v0, <4 x i32> %v1, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; NEON-NEXT:       [[TMP3:%.*]] = bitcast <8 x i32>* %ptr to <4 x i32>*
+; NEON-NEXT:       call void @llvm.aarch64.neon.st2.v4i32.p0v4i32(<4 x i32> [[TMP1]], <4 x i32> [[TMP2]], <4 x i32>* [[TMP3]])
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_undef_mask_factor2(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <4 x i32> %v0, <4 x i32> %v1, <8 x i32> <i32 undef, i32 undef, i32 undef, i32 undef, i32 2, i32 6, i32 3, i32 7>
+  store <8 x i32> %interleaved.vec, <8 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_undef_mask_factor3(<12 x i32>* %ptr, <4 x i32> %v0, <4 x i32> %v1, <4 x i32> %v2) {
+; NEON-LABEL:    @store_undef_mask_factor3(
+; NEON:            [[TMP1:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 8, i32 9, i32 10, i32 11>
+; NEON-NEXT:       [[TMP4:%.*]] = bitcast <12 x i32>* %ptr to <4 x i32>*
+; NEON-NEXT:       call void @llvm.aarch64.neon.st3.v4i32.p0v4i32(<4 x i32> [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> [[TMP3]], <4 x i32>* [[TMP4]])
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_undef_mask_factor3(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %s0 = shufflevector <4 x i32> %v0, <4 x i32> %v1, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %s1 = shufflevector <4 x i32> %v2, <4 x i32> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef>
+  %interleaved.vec = shufflevector <8 x i32> %s0, <8 x i32> %s1, <12 x i32> <i32 0, i32 4, i32 undef, i32 1, i32 undef, i32 9, i32 2, i32 6, i32 10, i32 3, i32 7, i32 11>
+  store <12 x i32> %interleaved.vec, <12 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_undef_mask_factor4(<16 x i32>* %ptr, <4 x i32> %v0, <4 x i32> %v1, <4 x i32> %v2, <4 x i32> %v3) {
+; NEON-LABEL:    @store_undef_mask_factor4(
+; NEON:            [[TMP1:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 8, i32 9, i32 10, i32 11>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 12, i32 13, i32 14, i32 15>
+; NEON-NEXT:       [[TMP5:%.*]] = bitcast <16 x i32>* %ptr to <4 x i32>*
+; NEON-NEXT:       call void @llvm.aarch64.neon.st4.v4i32.p0v4i32(<4 x i32> [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> [[TMP3]], <4 x i32> [[TMP4]], <4 x i32>* [[TMP5]])
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_undef_mask_factor4(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %s0 = shufflevector <4 x i32> %v0, <4 x i32> %v1, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %s1 = shufflevector <4 x i32> %v2, <4 x i32> %v3, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %interleaved.vec = shufflevector <8 x i32> %s0, <8 x i32> %s1, <16 x i32> <i32 0, i32 4, i32 8, i32 undef, i32 undef, i32 5, i32 9, i32 13, i32 2, i32 6, i32 10, i32 14, i32 3, i32 7, i32 11, i32 15>
+  store <16 x i32> %interleaved.vec, <16 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @load_illegal_factor2(<3 x float>* %ptr) nounwind {
+; NEON-LABEL:    @load_illegal_factor2(
+; NEON-NOT:        @llvm.aarch64.neon
+; NEON:            ret void
+; NO_NEON-LABEL: @load_illegal_factor2(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <3 x float>, <3 x float>* %ptr, align 16
+  %v0 = shufflevector <3 x float> %interleaved.vec, <3 x float> undef, <3 x i32> <i32 0, i32 2, i32 undef>
+  ret void
+}
+
+define void @store_illegal_factor2(<3 x float>* %ptr, <3 x float> %v0) nounwind {
+; NEON-LABEL:    @store_illegal_factor2(
+; NEON-NOT:        @llvm.aarch64.neon
+; NEON:            ret void
+; NO_NEON-LABEL: @store_illegal_factor2(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <3 x float> %v0, <3 x float> undef, <3 x i32> <i32 0, i32 2, i32 undef>
+  store <3 x float> %interleaved.vec, <3 x float>* %ptr, align 16
+  ret void
+}
+
+define void @store_general_mask_factor4(<8 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor4(
+; NEON-NEXT:       [[TMP1:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 4, i32 5>
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 16, i32 17>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 32, i32 33>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 8, i32 9>
+; NEON-NEXT:       [[TMP5:%.*]] = bitcast <8 x i32>* %ptr to <2 x i32>*
+; NEON-NEXT:       call void @llvm.aarch64.neon.st4.v2i32.p0v2i32(<2 x i32> [[TMP1]], <2 x i32> [[TMP2]], <2 x i32> [[TMP3]], <2 x i32> [[TMP4]], <2 x i32>* [[TMP5]])
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_general_mask_factor4(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <8 x i32> <i32 4, i32 16, i32 32, i32 8, i32 5, i32 17, i32 33, i32 9>
+  store <8 x i32> %interleaved.vec, <8 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor4_undefbeg(<8 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor4_undefbeg(
+; NEON-NEXT:       [[TMP1:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 4, i32 5>
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 16, i32 17>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 32, i32 33>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 8, i32 9>
+; NEON-NEXT:       [[TMP5:%.*]] = bitcast <8 x i32>* %ptr to <2 x i32>*
+; NEON-NEXT:       call void @llvm.aarch64.neon.st4.v2i32.p0v2i32(<2 x i32> [[TMP1]], <2 x i32> [[TMP2]], <2 x i32> [[TMP3]], <2 x i32> [[TMP4]], <2 x i32>* [[TMP5]])
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_general_mask_factor4_undefbeg(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <8 x i32> <i32 undef, i32 16, i32 32, i32 8, i32 5, i32 17, i32 33, i32 9>
+  store <8 x i32> %interleaved.vec, <8 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor4_undefend(<8 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor4_undefend(
+; NEON-NEXT:       [[TMP1:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 4, i32 5>
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 16, i32 17>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 32, i32 33>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 8, i32 9>
+; NEON-NEXT:       [[TMP5:%.*]] = bitcast <8 x i32>* %ptr to <2 x i32>*
+; NEON-NEXT:       call void @llvm.aarch64.neon.st4.v2i32.p0v2i32(<2 x i32> [[TMP1]], <2 x i32> [[TMP2]], <2 x i32> [[TMP3]], <2 x i32> [[TMP4]], <2 x i32>* [[TMP5]])
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_general_mask_factor4_undefend(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <8 x i32> <i32 4, i32 16, i32 32, i32 8, i32 5, i32 17, i32 33, i32 undef>
+  store <8 x i32> %interleaved.vec, <8 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor4_undefmid(<8 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor4_undefmid(
+; NEON-NEXT:       [[TMP1:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 4, i32 5>
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 16, i32 17>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 32, i32 33>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 8, i32 9>
+; NEON-NEXT:       [[TMP5:%.*]] = bitcast <8 x i32>* %ptr to <2 x i32>*
+; NEON-NEXT:       call void @llvm.aarch64.neon.st4.v2i32.p0v2i32(<2 x i32> [[TMP1]], <2 x i32> [[TMP2]], <2 x i32> [[TMP3]], <2 x i32> [[TMP4]], <2 x i32>* [[TMP5]])
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_general_mask_factor4_undefmid(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <8 x i32> <i32 4, i32 undef, i32 32, i32 8, i32 5, i32 17, i32 undef, i32 9>
+  store <8 x i32> %interleaved.vec, <8 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor4_undefmulti(<8 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor4_undefmulti(
+; NEON-NEXT:       [[TMP1:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 4, i32 5>
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 0, i32 1>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 0, i32 1>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 8, i32 9>
+; NEON-NEXT:       [[TMP5:%.*]] = bitcast <8 x i32>* %ptr to <2 x i32>*
+; NEON-NEXT:       call void @llvm.aarch64.neon.st4.v2i32.p0v2i32(<2 x i32> [[TMP1]], <2 x i32> [[TMP2]], <2 x i32> [[TMP3]], <2 x i32> [[TMP4]], <2 x i32>* [[TMP5]])
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_general_mask_factor4_undefmulti(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <8 x i32> <i32 4, i32 undef, i32 undef, i32 8, i32 undef, i32 undef, i32 undef, i32 9>
+  store <8 x i32> %interleaved.vec, <8 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor3(<12 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor3(
+; NEON-NEXT:       [[TMP1:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 32, i32 33, i32 34, i32 35>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 16, i32 17, i32 18, i32 19>
+; NEON-NEXT:       [[TMP4:%.*]] = bitcast <12 x i32>* %ptr to <4 x i32>*
+; NEON-NEXT:       call void @llvm.aarch64.neon.st3.v4i32.p0v4i32(<4 x i32> [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> [[TMP3]], <4 x i32>* [[TMP4]])
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_general_mask_factor3(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <12 x i32> <i32 4, i32 32, i32 16, i32 5, i32 33, i32 17, i32 6, i32 34, i32 18, i32 7, i32 35, i32 19>
+  store <12 x i32> %interleaved.vec, <12 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor3_undefmultimid(<12 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor3_undefmultimid(
+; NEON-NEXT:       [[TMP1:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 32, i32 33, i32 34, i32 35>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 16, i32 17, i32 18, i32 19>
+; NEON-NEXT:       [[TMP4:%.*]] = bitcast <12 x i32>* %ptr to <4 x i32>*
+; NEON-NEXT:       call void @llvm.aarch64.neon.st3.v4i32.p0v4i32(<4 x i32> [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> [[TMP3]], <4 x i32>* [[TMP4]])
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_general_mask_factor3_undefmultimid(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <12 x i32> <i32 4, i32 32, i32 16, i32 undef, i32 33, i32 17, i32 undef, i32 34, i32 18, i32 7, i32 35, i32 19>
+  store <12 x i32> %interleaved.vec, <12 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor3_undef_fail(<12 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor3_undef_fail(
+; NEON-NOT:        @llvm.aarch64.neon
+; NEON:            ret void
+; NO_NEON-LABEL: @store_general_mask_factor3_undef_fail(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <12 x i32> <i32 4, i32 32, i32 16, i32 undef, i32 33, i32 17, i32 undef, i32 34, i32 18, i32 8, i32 35, i32 19>
+  store <12 x i32> %interleaved.vec, <12 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor3_undeflane(<12 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor3_undeflane(
+; NEON-NEXT:       [[TMP1:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 32, i32 33, i32 34, i32 35>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 16, i32 17, i32 18, i32 19>
+; NEON-NEXT:       [[TMP4:%.*]] = bitcast <12 x i32>* %ptr to <4 x i32>*
+; NEON-NEXT:       call void @llvm.aarch64.neon.st3.v4i32.p0v4i32(<4 x i32> [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> [[TMP3]], <4 x i32>* [[TMP4]])
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_general_mask_factor3_undeflane(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <12 x i32> <i32 undef, i32 32, i32 16, i32 undef, i32 33, i32 17, i32 undef, i32 34, i32 18, i32 undef, i32 35, i32 19>
+  store <12 x i32> %interleaved.vec, <12 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor3_negativestart(<12 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor3_negativestart(
+; NEON-NOT:        @llvm.aarch64.neon
+; NEON:            ret void
+; NO_NEON-LABEL: @store_general_mask_factor3_negativestart(
+; NO_NEON-NOT:     @llvm.aarch64.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <12 x i32> <i32 undef, i32 32, i32 16, i32 undef, i32 33, i32 17, i32 undef, i32 34, i32 18, i32 2, i32 35, i32 19>
+  store <12 x i32> %interleaved.vec, <12 x i32>* %ptr, align 4
+  ret void
+}
diff --git a/llvm/test/Transforms/InterleavedAccess/AArch64/lit.local.cfg b/llvm/test/Transforms/InterleavedAccess/AArch64/lit.local.cfg
new file mode 100644
index 00000000000..304f2436ece
--- /dev/null
+++ b/llvm/test/Transforms/InterleavedAccess/AArch64/lit.local.cfg
@@ -0,0 +1,2 @@
+if not 'AArch64' in config.root.targets:
+  config.unsupported = True
diff --git a/llvm/test/Transforms/InterleavedAccess/ARM/interleaved-accesses-extract-user.ll b/llvm/test/Transforms/InterleavedAccess/ARM/interleaved-accesses-extract-user.ll
new file mode 100644
index 00000000000..2ea92511749
--- /dev/null
+++ b/llvm/test/Transforms/InterleavedAccess/ARM/interleaved-accesses-extract-user.ll
@@ -0,0 +1,113 @@
+; RUN: opt < %s -mattr=+neon -interleaved-access -S | FileCheck %s
+
+target datalayout = "e-m:e-p:32:32-i64:64-v128:64:128-n32-S64"
+target triple = "arm---eabi"
+
+define void @extract_user_basic(<8 x i32>* %ptr, i1 %c) {
+; CHECK-LABEL: @extract_user_basic(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <8 x i32>* %ptr to i8*
+; CHECK-NEXT:    [[VLDN:%.*]] = call { <4 x i32>, <4 x i32> } @llvm.arm.neon.vld2.v4i32.p0i8(i8* [[TMP0]], i32 8)
+; CHECK-NEXT:    [[TMP1:%.*]] = extractvalue { <4 x i32>, <4 x i32> } [[VLDN]], 0
+; CHECK-NEXT:    br i1 %c, label %if.then, label %if.merge
+; CHECK:       if.then:
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <4 x i32> [[TMP1]], i64 1
+; CHECK-NEXT:    br label %if.merge
+; CHECK:       if.merge:
+; CHECK-NEXT:    ret void
+;
+entry:
+  %interleaved.vec = load <8 x i32>, <8 x i32>* %ptr, align 8
+  %v0 = shufflevector <8 x i32> %interleaved.vec, <8 x i32> undef, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
+  br i1 %c, label %if.then, label %if.merge
+
+if.then:
+  %e0 = extractelement <8 x i32> %interleaved.vec, i32 2
+  br label %if.merge
+
+if.merge:
+  ret void
+}
+
+define void @extract_user_multi(<8 x i32>* %ptr, i1 %c) {
+; CHECK-LABEL: @extract_user_multi(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <8 x i32>* %ptr to i8*
+; CHECK-NEXT:    [[VLDN:%.*]] = call { <4 x i32>, <4 x i32> } @llvm.arm.neon.vld2.v4i32.p0i8(i8* [[TMP0]], i32 8)
+; CHECK-NEXT:    [[TMP1:%.*]] = extractvalue { <4 x i32>, <4 x i32> } [[VLDN]], 0
+; CHECK-NEXT:    br i1 %c, label %if.then, label %if.merge
+; CHECK:       if.then:
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <4 x i32> [[TMP1]], i64 0
+; CHECK-NEXT:    br label %if.merge
+; CHECK:       if.merge:
+; CHECK-NEXT:    [[TMP3:%.*]] = extractelement <4 x i32> [[TMP1]], i64 1
+; CHECK-NEXT:    ret void
+;
+entry:
+  %interleaved.vec = load <8 x i32>, <8 x i32>* %ptr, align 8
+  %v0 = shufflevector <8 x i32> %interleaved.vec, <8 x i32> undef, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
+  br i1 %c, label %if.then, label %if.merge
+
+if.then:
+  %e0 = extractelement <8 x i32> %interleaved.vec, i32 0
+  br label %if.merge
+
+if.merge:
+  %e1 = extractelement <8 x i32> %interleaved.vec, i32 2
+  ret void
+}
+
+define void @extract_user_multi_no_dom(<8 x i32>* %ptr, i1 %c) {
+; CHECK-LABEL: @extract_user_multi_no_dom(
+; CHECK-NOT:     @llvm.arm.neon
+; CHECK:         ret void
+;
+entry:
+  %interleaved.vec = load <8 x i32>, <8 x i32>* %ptr, align 8
+  %e0 = extractelement <8 x i32> %interleaved.vec, i32 0
+  br i1 %c, label %if.then, label %if.merge
+
+if.then:
+  %v0 = shufflevector <8 x i32> %interleaved.vec, <8 x i32> undef, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
+  %e1 = extractelement <8 x i32> %interleaved.vec, i32 2
+  br label %if.merge
+
+if.merge:
+  ret void
+}
+
+define void @extract_user_wrong_const_index(<8 x i32>* %ptr) {
+; CHECK-LABEL: @extract_user_wrong_const_index(
+; CHECK-NOT:     @llvm.arm.neon
+; CHECK:         ret void
+;
+entry:
+  %interleaved.vec = load <8 x i32>, <8 x i32>* %ptr, align 8
+  %v0 = shufflevector <8 x i32> %interleaved.vec, <8 x i32> undef, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
+  %e0 = extractelement <8 x i32> %interleaved.vec, i32 1
+  ret void
+}
+
+define void @extract_user_undef_index(<8 x i32>* %ptr) {
+; CHECK-LABEL: @extract_user_undef_index(
+; CHECK-NOT:     @llvm.arm.neon
+; CHECK:         ret void
+;
+entry:
+  %interleaved.vec = load <8 x i32>, <8 x i32>* %ptr, align 8
+  %v0 = shufflevector <8 x i32> %interleaved.vec, <8 x i32> undef, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
+  %e0 = extractelement <8 x i32> %interleaved.vec, i32 undef
+  ret void
+}
+
+define void @extract_user_var_index(<8 x i32>* %ptr, i32 %i) {
+; CHECK-LABEL: @extract_user_var_index(
+; CHECK-NOT:     @llvm.arm.neon
+; CHECK:         ret void
+;
+entry:
+  %interleaved.vec = load <8 x i32>, <8 x i32>* %ptr, align 8
+  %v0 = shufflevector <8 x i32> %interleaved.vec, <8 x i32> undef, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
+  %e0 = extractelement <8 x i32> %interleaved.vec, i32 %i
+  ret void
+}
diff --git a/llvm/test/Transforms/InterleavedAccess/ARM/interleaved-accesses.ll b/llvm/test/Transforms/InterleavedAccess/ARM/interleaved-accesses.ll
new file mode 100644
index 00000000000..caaaa21c5a1
--- /dev/null
+++ b/llvm/test/Transforms/InterleavedAccess/ARM/interleaved-accesses.ll
@@ -0,0 +1,628 @@
+; RUN: opt < %s -mattr=+neon -interleaved-access -S | FileCheck %s -check-prefix=NEON
+; RUN: opt < %s -interleaved-access -S | FileCheck %s -check-prefix=NO_NEON
+
+target datalayout = "e-m:e-p:32:32-i64:64-v128:64:128-n32-S64"
+target triple = "arm---eabi"
+
+define void @load_factor2(<16 x i8>* %ptr) {
+; NEON-LABEL:    @load_factor2(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <16 x i8>* %ptr to i8*
+; NEON-NEXT:       [[VLDN:%.*]] = call { <8 x i8>, <8 x i8> } @llvm.arm.neon.vld2.v8i8.p0i8(i8* [[TMP1]], i32 4)
+; NEON-NEXT:       [[TMP2:%.*]] = extractvalue { <8 x i8>, <8 x i8> } [[VLDN]], 1
+; NEON-NEXT:       [[TMP3:%.*]] = extractvalue { <8 x i8>, <8 x i8> } [[VLDN]], 0
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @load_factor2(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <16 x i8>, <16 x i8>* %ptr, align 4
+  %v0 = shufflevector <16 x i8> %interleaved.vec, <16 x i8> undef, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14>
+  %v1 = shufflevector <16 x i8> %interleaved.vec, <16 x i8> undef, <8 x i32> <i32 1, i32 3, i32 5, i32 7, i32 9, i32 11, i32 13, i32 15>
+  ret void
+}
+
+define void @load_factor3(<6 x i32>* %ptr) {
+; NEON-LABEL:    @load_factor3(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <6 x i32>* %ptr to i8*
+; NEON-NEXT:       [[VLDN:%.*]] = call { <2 x i32>, <2 x i32>, <2 x i32> } @llvm.arm.neon.vld3.v2i32.p0i8(i8* [[TMP1]], i32 4)
+; NEON-NEXT:       [[TMP2:%.*]] = extractvalue { <2 x i32>, <2 x i32>, <2 x i32> } [[VLDN]], 2
+; NEON-NEXT:       [[TMP3:%.*]] = extractvalue { <2 x i32>, <2 x i32>, <2 x i32> } [[VLDN]], 1
+; NEON-NEXT:       [[TMP4:%.*]] = extractvalue { <2 x i32>, <2 x i32>, <2 x i32> } [[VLDN]], 0
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @load_factor3(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <6 x i32>, <6 x i32>* %ptr, align 4
+  %v0 = shufflevector <6 x i32> %interleaved.vec, <6 x i32> undef, <2 x i32> <i32 0, i32 3>
+  %v1 = shufflevector <6 x i32> %interleaved.vec, <6 x i32> undef, <2 x i32> <i32 1, i32 4>
+  %v2 = shufflevector <6 x i32> %interleaved.vec, <6 x i32> undef, <2 x i32> <i32 2, i32 5>
+  ret void
+}
+
+define void @load_factor4(<16 x i32>* %ptr) {
+; NEON-LABEL:    @load_factor4(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <16 x i32>* %ptr to i8*
+; NEON-NEXT:       [[VLDN:%.*]] = call { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } @llvm.arm.neon.vld4.v4i32.p0i8(i8* [[TMP1]], i32 4)
+; NEON-NEXT:       [[TMP2:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[VLDN]], 3
+; NEON-NEXT:       [[TMP3:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[VLDN]], 2
+; NEON-NEXT:       [[TMP4:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[VLDN]], 1
+; NEON-NEXT:       [[TMP5:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[VLDN]], 0
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @load_factor4(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <16 x i32>, <16 x i32>* %ptr, align 4
+  %v0 = shufflevector <16 x i32> %interleaved.vec, <16 x i32> undef, <4 x i32> <i32 0, i32 4, i32 8, i32 12>
+  %v1 = shufflevector <16 x i32> %interleaved.vec, <16 x i32> undef, <4 x i32> <i32 1, i32 5, i32 9, i32 13>
+  %v2 = shufflevector <16 x i32> %interleaved.vec, <16 x i32> undef, <4 x i32> <i32 2, i32 6, i32 10, i32 14>
+  %v3 = shufflevector <16 x i32> %interleaved.vec, <16 x i32> undef, <4 x i32> <i32 3, i32 7, i32 11, i32 15>
+  ret void
+}
+
+define void @store_factor2(<16 x i8>* %ptr, <8 x i8> %v0, <8 x i8> %v1) {
+; NEON-LABEL:    @store_factor2(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <16 x i8>* %ptr to i8*
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <8 x i8> %v0, <8 x i8> %v1, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <8 x i8> %v0, <8 x i8> %v1, <8 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
+; NEON-NEXT:       call void @llvm.arm.neon.vst2.p0i8.v8i8(i8* [[TMP1]], <8 x i8> [[TMP2]], <8 x i8> [[TMP3]], i32 4)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_factor2(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <8 x i8> %v0, <8 x i8> %v1, <16 x i32> <i32 0, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11, i32 4, i32 12, i32 5, i32 13, i32 6, i32 14, i32 7, i32 15>
+  store <16 x i8> %interleaved.vec, <16 x i8>* %ptr, align 4
+  ret void
+}
+
+define void @store_factor3(<12 x i32>* %ptr, <4 x i32> %v0, <4 x i32> %v1, <4 x i32> %v2) {
+; NEON-LABEL:    @store_factor3(
+; NEON:            [[TMP1:%.*]] = bitcast <12 x i32>* %ptr to i8*
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 8, i32 9, i32 10, i32 11>
+; NEON-NEXT:       call void @llvm.arm.neon.vst3.p0i8.v4i32(i8* [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> [[TMP3]], <4 x i32> [[TMP4]], i32 4)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_factor3(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %s0 = shufflevector <4 x i32> %v0, <4 x i32> %v1, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %s1 = shufflevector <4 x i32> %v2, <4 x i32> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef>
+  %interleaved.vec = shufflevector <8 x i32> %s0, <8 x i32> %s1, <12 x i32> <i32 0, i32 4, i32 8, i32 1, i32 5, i32 9, i32 2, i32 6, i32 10, i32 3, i32 7, i32 11>
+  store <12 x i32> %interleaved.vec, <12 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_factor4(<16 x i32>* %ptr, <4 x i32> %v0, <4 x i32> %v1, <4 x i32> %v2, <4 x i32> %v3) {
+; NEON-LABEL:    @store_factor4(
+; NEON:            [[TMP1:%.*]] = bitcast <16 x i32>* %ptr to i8*
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 8, i32 9, i32 10, i32 11>
+; NEON-NEXT:       [[TMP5:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 12, i32 13, i32 14, i32 15>
+; NEON-NEXT:       call void @llvm.arm.neon.vst4.p0i8.v4i32(i8* [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> [[TMP3]], <4 x i32> [[TMP4]], <4 x i32> [[TMP5]], i32 4)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_factor4(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %s0 = shufflevector <4 x i32> %v0, <4 x i32> %v1, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %s1 = shufflevector <4 x i32> %v2, <4 x i32> %v3, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %interleaved.vec = shufflevector <8 x i32> %s0, <8 x i32> %s1, <16 x i32> <i32 0, i32 4, i32 8, i32 12, i32 1, i32 5, i32 9, i32 13, i32 2, i32 6, i32 10, i32 14, i32 3, i32 7, i32 11, i32 15>
+  store <16 x i32> %interleaved.vec, <16 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @load_ptrvec_factor2(<4 x i32*>* %ptr) {
+; NEON-LABEL:    @load_ptrvec_factor2(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <4 x i32*>* %ptr to i8*
+; NEON-NEXT:       [[VLDN:%.*]] = call { <2 x i32>, <2 x i32> } @llvm.arm.neon.vld2.v2i32.p0i8(i8* [[TMP1]], i32 4)
+; NEON-NEXT:       [[TMP2:%.*]] = extractvalue { <2 x i32>, <2 x i32> } [[VLDN]], 0
+; NEON-NEXT:       [[TMP3:%.*]] = inttoptr <2 x i32> [[TMP2]] to <2 x i32*>
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @load_ptrvec_factor2(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <4 x i32*>, <4 x i32*>* %ptr, align 4
+  %v0 = shufflevector <4 x i32*> %interleaved.vec, <4 x i32*> undef, <2 x i32> <i32 0, i32 2>
+  ret void
+}
+
+define void @load_ptrvec_factor3(<6 x i32*>* %ptr) {
+; NEON-LABEL:    @load_ptrvec_factor3(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <6 x i32*>* %ptr to i8*
+; NEON-NEXT:       [[VLDN:%.*]] = call { <2 x i32>, <2 x i32>, <2 x i32> } @llvm.arm.neon.vld3.v2i32.p0i8(i8* [[TMP1]], i32 4)
+; NEON-NEXT:       [[TMP2:%.*]] = extractvalue { <2 x i32>, <2 x i32>, <2 x i32> } [[VLDN]], 2
+; NEON-NEXT:       [[TMP3:%.*]] = inttoptr <2 x i32> [[TMP2]] to <2 x i32*>
+; NEON-NEXT:       [[TMP4:%.*]] = extractvalue { <2 x i32>, <2 x i32>, <2 x i32> } [[VLDN]], 1
+; NEON-NEXT:       [[TMP5:%.*]] = inttoptr <2 x i32> [[TMP4]] to <2 x i32*>
+; NEON-NEXT:       [[TMP6:%.*]] = extractvalue { <2 x i32>, <2 x i32>, <2 x i32> } [[VLDN]], 0
+; NEON-NEXT:       [[TMP7:%.*]] = inttoptr <2 x i32> [[TMP6]] to <2 x i32*>
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @load_ptrvec_factor3(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <6 x i32*>, <6 x i32*>* %ptr, align 4
+  %v0 = shufflevector <6 x i32*> %interleaved.vec, <6 x i32*> undef, <2 x i32> <i32 0, i32 3>
+  %v1 = shufflevector <6 x i32*> %interleaved.vec, <6 x i32*> undef, <2 x i32> <i32 1, i32 4>
+  %v2 = shufflevector <6 x i32*> %interleaved.vec, <6 x i32*> undef, <2 x i32> <i32 2, i32 5>
+  ret void
+}
+
+define void @load_ptrvec_factor4(<8 x i32*>* %ptr) {
+; NEON-LABEL:    @load_ptrvec_factor4(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <8 x i32*>* %ptr to i8*
+; NEON-NEXT:       [[VLDN:%.*]] = call { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> } @llvm.arm.neon.vld4.v2i32.p0i8(i8* [[TMP1]], i32 4)
+; NEON-NEXT:       [[TMP2:%.*]] = extractvalue { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> } [[VLDN]], 3
+; NEON-NEXT:       [[TMP3:%.*]] = inttoptr <2 x i32> [[TMP2]] to <2 x i32*>
+; NEON-NEXT:       [[TMP4:%.*]] = extractvalue { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> } [[VLDN]], 2
+; NEON-NEXT:       [[TMP5:%.*]] = inttoptr <2 x i32> [[TMP4]] to <2 x i32*>
+; NEON-NEXT:       [[TMP6:%.*]] = extractvalue { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> } [[VLDN]], 1
+; NEON-NEXT:       [[TMP7:%.*]] = inttoptr <2 x i32> [[TMP6]] to <2 x i32*>
+; NEON-NEXT:       [[TMP8:%.*]] = extractvalue { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> } [[VLDN]], 0
+; NEON-NEXT:       [[TMP9:%.*]] = inttoptr <2 x i32> [[TMP8]] to <2 x i32*>
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @load_ptrvec_factor4(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <8 x i32*>, <8 x i32*>* %ptr, align 4
+  %v0 = shufflevector <8 x i32*> %interleaved.vec, <8 x i32*> undef, <2 x i32> <i32 0, i32 4>
+  %v1 = shufflevector <8 x i32*> %interleaved.vec, <8 x i32*> undef, <2 x i32> <i32 1, i32 5>
+  %v2 = shufflevector <8 x i32*> %interleaved.vec, <8 x i32*> undef, <2 x i32> <i32 2, i32 6>
+  %v3 = shufflevector <8 x i32*> %interleaved.vec, <8 x i32*> undef, <2 x i32> <i32 3, i32 7>
+  ret void
+}
+
+define void @store_ptrvec_factor2(<4 x i32*>* %ptr, <2 x i32*> %v0, <2 x i32*> %v1) {
+; NEON-LABEL:    @store_ptrvec_factor2(
+; NEON-NEXT:       [[TMP1:%.*]] = ptrtoint <2 x i32*> %v0 to <2 x i32>
+; NEON-NEXT:       [[TMP2:%.*]] = ptrtoint <2 x i32*> %v1 to <2 x i32>
+; NEON-NEXT:       [[TMP3:%.*]] = bitcast <4 x i32*>* %ptr to i8*
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <2 x i32> [[TMP1]], <2 x i32> [[TMP2]], <2 x i32> <i32 0, i32 1>
+; NEON-NEXT:       [[TMP5:%.*]] = shufflevector <2 x i32> [[TMP1]], <2 x i32> [[TMP2]], <2 x i32> <i32 2, i32 3>
+; NEON-NEXT:       call void @llvm.arm.neon.vst2.p0i8.v2i32(i8* [[TMP3]], <2 x i32> [[TMP4]], <2 x i32> [[TMP5]], i32 4)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_ptrvec_factor2(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <2 x i32*> %v0, <2 x i32*> %v1, <4 x i32> <i32 0, i32 2, i32 1, i32 3>
+  store <4 x i32*> %interleaved.vec, <4 x i32*>* %ptr, align 4
+  ret void
+}
+
+define void @store_ptrvec_factor3(<6 x i32*>* %ptr, <2 x i32*> %v0, <2 x i32*> %v1, <2 x i32*> %v2) {
+; NEON-LABEL:    @store_ptrvec_factor3(
+; NEON:            [[TMP1:%.*]] = ptrtoint <4 x i32*> %s0 to <4 x i32>
+; NEON-NEXT:       [[TMP2:%.*]] = ptrtoint <4 x i32*> %s1 to <4 x i32>
+; NEON-NEXT:       [[TMP3:%.*]] = bitcast <6 x i32*>* %ptr to i8*
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> [[TMP2]], <2 x i32> <i32 0, i32 1>
+; NEON-NEXT:       [[TMP5:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> [[TMP2]], <2 x i32> <i32 2, i32 3>
+; NEON-NEXT:       [[TMP6:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> [[TMP2]], <2 x i32> <i32 4, i32 5>
+; NEON-NEXT:       call void @llvm.arm.neon.vst3.p0i8.v2i32(i8* [[TMP3]], <2 x i32> [[TMP4]], <2 x i32> [[TMP5]], <2 x i32> [[TMP6]], i32 4)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_ptrvec_factor3(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %s0 = shufflevector <2 x i32*> %v0, <2 x i32*> %v1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+  %s1 = shufflevector <2 x i32*> %v2, <2 x i32*> undef, <4 x i32> <i32 0, i32 1, i32 undef, i32 undef>
+  %interleaved.vec = shufflevector <4 x i32*> %s0, <4 x i32*> %s1, <6 x i32> <i32 0, i32 2, i32 4, i32 1, i32 3, i32 5>
+  store <6 x i32*> %interleaved.vec, <6 x i32*>* %ptr, align 4
+  ret void
+}
+
+define void @store_ptrvec_factor4(<8 x i32*>* %ptr, <2 x i32*> %v0, <2 x i32*> %v1, <2 x i32*> %v2, <2 x i32*> %v3) {
+; NEON-LABEL:    @store_ptrvec_factor4(
+; NEON:            [[TMP1:%.*]] = ptrtoint <4 x i32*> %s0 to <4 x i32>
+; NEON-NEXT:       [[TMP2:%.*]] = ptrtoint <4 x i32*> %s1 to <4 x i32>
+; NEON-NEXT:       [[TMP3:%.*]] = bitcast <8 x i32*>* %ptr to i8*
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> [[TMP2]], <2 x i32> <i32 0, i32 1>
+; NEON-NEXT:       [[TMP5:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> [[TMP2]], <2 x i32> <i32 2, i32 3>
+; NEON-NEXT:       [[TMP6:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> [[TMP2]], <2 x i32> <i32 4, i32 5>
+; NEON-NEXT:       [[TMP7:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> [[TMP2]], <2 x i32> <i32 6, i32 7>
+; NEON-NEXT:       call void @llvm.arm.neon.vst4.p0i8.v2i32(i8* [[TMP3]], <2 x i32> [[TMP4]], <2 x i32> [[TMP5]], <2 x i32> [[TMP6]], <2 x i32> [[TMP7]], i32 4)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_ptrvec_factor4(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %s0 = shufflevector <2 x i32*> %v0, <2 x i32*> %v1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+  %s1 = shufflevector <2 x i32*> %v2, <2 x i32*> %v3, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+  %interleaved.vec = shufflevector <4 x i32*> %s0, <4 x i32*> %s1, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 1, i32 3, i32 5, i32 7>
+  store <8 x i32*> %interleaved.vec, <8 x i32*>* %ptr, align 4
+  ret void
+}
+
+define void @load_undef_mask_factor2(<8 x i32>* %ptr) {
+; NEON-LABEL:    @load_undef_mask_factor2(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <8 x i32>* %ptr to i8*
+; NEON-NEXT:       [[VLDN:%.*]] = call { <4 x i32>, <4 x i32> } @llvm.arm.neon.vld2.v4i32.p0i8(i8* [[TMP1]], i32 4)
+; NEON-NEXT:       [[TMP2:%.*]] = extractvalue { <4 x i32>, <4 x i32> } [[VLDN]], 1
+; NEON-NEXT:       [[TMP3:%.*]] = extractvalue { <4 x i32>, <4 x i32> } [[VLDN]], 0
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @load_undef_mask_factor2(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <8 x i32>, <8 x i32>* %ptr, align 4
+  %v0 = shufflevector <8 x i32> %interleaved.vec, <8 x i32> undef, <4 x i32> <i32 undef, i32 2, i32 undef, i32 6>
+  %v1 = shufflevector <8 x i32> %interleaved.vec, <8 x i32> undef, <4 x i32> <i32 undef, i32 3, i32 undef, i32 7>
+  ret void
+}
+
+define void @load_undef_mask_factor3(<12 x i32>* %ptr) {
+; NEON-LABEL:    @load_undef_mask_factor3(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <12 x i32>* %ptr to i8*
+; NEON-NEXT:       [[VLDN:%.*]] = call { <4 x i32>, <4 x i32>, <4 x i32> } @llvm.arm.neon.vld3.v4i32.p0i8(i8* [[TMP1]], i32 4)
+; NEON-NEXT:       [[TMP2:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32> } [[VLDN]], 2
+; NEON-NEXT:       [[TMP3:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32> } [[VLDN]], 1
+; NEON-NEXT:       [[TMP4:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32> } [[VLDN]], 0
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @load_undef_mask_factor3(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <12 x i32>, <12 x i32>* %ptr, align 4
+  %v0 = shufflevector <12 x i32> %interleaved.vec, <12 x i32> undef, <4 x i32> <i32 0, i32 3, i32 6, i32 9>
+  %v1 = shufflevector <12 x i32> %interleaved.vec, <12 x i32> undef, <4 x i32> <i32 1, i32 4, i32 7, i32 10>
+  %v2 = shufflevector <12 x i32> %interleaved.vec, <12 x i32> undef, <4 x i32> <i32 2, i32 undef, i32 undef, i32 undef>
+  ret void
+}
+
+define void @load_undef_mask_factor4(<16 x i32>* %ptr) {
+; NEON-LABEL:    @load_undef_mask_factor4(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <16 x i32>* %ptr to i8*
+; NEON-NEXT:       [[VLDN:%.*]] = call { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } @llvm.arm.neon.vld4.v4i32.p0i8(i8* [[TMP1]], i32 4)
+; NEON-NEXT:       [[TMP2:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[VLDN]], 3
+; NEON-NEXT:       [[TMP3:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[VLDN]], 2
+; NEON-NEXT:       [[TMP4:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[VLDN]], 1
+; NEON-NEXT:       [[TMP5:%.*]] = extractvalue { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[VLDN]], 0
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @load_undef_mask_factor4(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <16 x i32>, <16 x i32>* %ptr, align 4
+  %v0 = shufflevector <16 x i32> %interleaved.vec, <16 x i32> undef, <4 x i32> <i32 0, i32 4, i32 undef, i32 undef>
+  %v1 = shufflevector <16 x i32> %interleaved.vec, <16 x i32> undef, <4 x i32> <i32 1, i32 5, i32 undef, i32 undef>
+  %v2 = shufflevector <16 x i32> %interleaved.vec, <16 x i32> undef, <4 x i32> <i32 2, i32 6, i32 undef, i32 undef>
+  %v3 = shufflevector <16 x i32> %interleaved.vec, <16 x i32> undef, <4 x i32> <i32 3, i32 7, i32 undef, i32 undef>
+  ret void
+}
+
+define void @store_undef_mask_factor2(<8 x i32>* %ptr, <4 x i32> %v0, <4 x i32> %v1) {
+; NEON-LABEL:    @store_undef_mask_factor2(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <8 x i32>* %ptr to i8*
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <4 x i32> %v0, <4 x i32> %v1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <4 x i32> %v0, <4 x i32> %v1, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; NEON-NEXT:       call void @llvm.arm.neon.vst2.p0i8.v4i32(i8* [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> [[TMP3]], i32 4)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_undef_mask_factor2(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <4 x i32> %v0, <4 x i32> %v1, <8 x i32> <i32 undef, i32 undef, i32 undef, i32 undef, i32 2, i32 6, i32 3, i32 7>
+  store <8 x i32> %interleaved.vec, <8 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_undef_mask_factor3(<12 x i32>* %ptr, <4 x i32> %v0, <4 x i32> %v1, <4 x i32> %v2) {
+; NEON-LABEL:    @store_undef_mask_factor3(
+; NEON:            [[TMP1:%.*]] = bitcast <12 x i32>* %ptr to i8*
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 8, i32 9, i32 10, i32 11>
+; NEON-NEXT:       call void @llvm.arm.neon.vst3.p0i8.v4i32(i8* [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> [[TMP3]], <4 x i32> [[TMP4]], i32 4)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_undef_mask_factor3(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %s0 = shufflevector <4 x i32> %v0, <4 x i32> %v1, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %s1 = shufflevector <4 x i32> %v2, <4 x i32> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef>
+  %interleaved.vec = shufflevector <8 x i32> %s0, <8 x i32> %s1, <12 x i32> <i32 0, i32 4, i32 undef, i32 1, i32 undef, i32 9, i32 2, i32 6, i32 10, i32 3, i32 7, i32 11>
+  store <12 x i32> %interleaved.vec, <12 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_undef_mask_factor4(<16 x i32>* %ptr, <4 x i32> %v0, <4 x i32> %v1, <4 x i32> %v2, <4 x i32> %v3) {
+; NEON-LABEL:    @store_undef_mask_factor4(
+; NEON:            [[TMP1:%.*]] = bitcast <16 x i32>* %ptr to i8*
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 8, i32 9, i32 10, i32 11>
+; NEON-NEXT:       [[TMP5:%.*]] = shufflevector <8 x i32> %s0, <8 x i32> %s1, <4 x i32> <i32 12, i32 13, i32 14, i32 15>
+; NEON-NEXT:       call void @llvm.arm.neon.vst4.p0i8.v4i32(i8* [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> [[TMP3]], <4 x i32> [[TMP4]], <4 x i32> [[TMP5]], i32 4)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_undef_mask_factor4(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %s0 = shufflevector <4 x i32> %v0, <4 x i32> %v1, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %s1 = shufflevector <4 x i32> %v2, <4 x i32> %v3, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %interleaved.vec = shufflevector <8 x i32> %s0, <8 x i32> %s1, <16 x i32> <i32 0, i32 4, i32 8, i32 undef, i32 undef, i32 5, i32 9, i32 13, i32 2, i32 6, i32 10, i32 14, i32 3, i32 7, i32 11, i32 15>
+  store <16 x i32> %interleaved.vec, <16 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @load_address_space(<4 x i32> addrspace(1)* %ptr) {
+; NEON-LABEL:    @load_address_space(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <4 x i32> addrspace(1)* %ptr to i8 addrspace(1)*
+; NEON-NEXT:       [[VLDN:%.*]] = call { <2 x i32>, <2 x i32>, <2 x i32> } @llvm.arm.neon.vld3.v2i32.p1i8(i8 addrspace(1)* [[TMP1]], i32 0)
+; NEON-NEXT:       [[TMP2:%.*]] = extractvalue { <2 x i32>, <2 x i32>, <2 x i32> } [[VLDN]], 2
+; NEON-NEXT:       [[TMP3:%.*]] = extractvalue { <2 x i32>, <2 x i32>, <2 x i32> } [[VLDN]], 1
+; NEON-NEXT:       [[TMP4:%.*]] = extractvalue { <2 x i32>, <2 x i32>, <2 x i32> } [[VLDN]], 0
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @load_address_space(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <4 x i32>, <4 x i32> addrspace(1)* %ptr
+  %v0 = shufflevector <4 x i32> %interleaved.vec, <4 x i32> undef, <2 x i32> <i32 0, i32 3>
+  %v1 = shufflevector <4 x i32> %interleaved.vec, <4 x i32> undef, <2 x i32> <i32 1, i32 4>
+  %v2 = shufflevector <4 x i32> %interleaved.vec, <4 x i32> undef, <2 x i32> <i32 2, i32 5>
+  ret void
+}
+
+define void @store_address_space(<4 x i32> addrspace(1)* %ptr, <2 x i32> %v0, <2 x i32> %v1) {
+; NEON-LABEL:    @store_address_space(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <4 x i32> addrspace(1)* %ptr to i8 addrspace(1)*
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <2 x i32> %v0, <2 x i32> %v1, <2 x i32> <i32 0, i32 1>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <2 x i32> %v0, <2 x i32> %v1, <2 x i32> <i32 2, i32 3>
+; NEON-NEXT:       call void @llvm.arm.neon.vst2.p1i8.v2i32(i8 addrspace(1)* [[TMP1]], <2 x i32> [[TMP2]], <2 x i32> [[TMP3]], i32 0)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_address_space(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <2 x i32> %v0, <2 x i32> %v1, <4 x i32> <i32 0, i32 2, i32 1, i32 3>
+  store <4 x i32> %interleaved.vec, <4 x i32> addrspace(1)* %ptr
+  ret void
+}
+
+define void @load_illegal_factor2(<3 x float>* %ptr) nounwind {
+; NEON-LABEL:    @load_illegal_factor2(
+; NEON-NOT:        @llvm.arm.neon
+; NEON:            ret void
+; NO_NEON-LABEL: @load_illegal_factor2(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = load <3 x float>, <3 x float>* %ptr, align 16
+  %v0 = shufflevector <3 x float> %interleaved.vec, <3 x float> undef, <3 x i32> <i32 0, i32 2, i32 undef>
+  ret void
+}
+
+define void @store_illegal_factor2(<3 x float>* %ptr, <3 x float> %v0) nounwind {
+; NEON-LABEL:    @store_illegal_factor2(
+; NEON-NOT:        @llvm.arm.neon
+; NEON:            ret void
+; NO_NEON-LABEL: @store_illegal_factor2(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <3 x float> %v0, <3 x float> undef, <3 x i32> <i32 0, i32 2, i32 undef>
+  store <3 x float> %interleaved.vec, <3 x float>* %ptr, align 16
+  ret void
+}
+
+define void @store_general_mask_factor4(<8 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor4(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <8 x i32>* %ptr to i8*
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 4, i32 5>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 16, i32 17>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 32, i32 33>
+; NEON-NEXT:       [[TMP5:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 8, i32 9>
+; NEON-NEXT:       call void @llvm.arm.neon.vst4.p0i8.v2i32(i8* [[TMP1]], <2 x i32> [[TMP2]], <2 x i32> [[TMP3]], <2 x i32> [[TMP4]], <2 x i32> [[TMP5]], i32 4)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_general_mask_factor4(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <8 x i32> <i32 4, i32 16, i32 32, i32 8, i32 5, i32 17, i32 33, i32 9>
+  store <8 x i32> %interleaved.vec, <8 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor4_undefbeg(<8 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor4_undefbeg(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <8 x i32>* %ptr to i8*
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 4, i32 5>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 16, i32 17>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 32, i32 33>
+; NEON-NEXT:       [[TMP5:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 8, i32 9>
+; NEON-NEXT:       call void @llvm.arm.neon.vst4.p0i8.v2i32(i8* [[TMP1]], <2 x i32> [[TMP2]], <2 x i32> [[TMP3]], <2 x i32> [[TMP4]], <2 x i32> [[TMP5]], i32 4)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_general_mask_factor4_undefbeg(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <8 x i32> <i32 undef, i32 16, i32 32, i32 8, i32 5, i32 17, i32 33, i32 9>
+  store <8 x i32> %interleaved.vec, <8 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor4_undefend(<8 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor4_undefend(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <8 x i32>* %ptr to i8*
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 4, i32 5>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 16, i32 17>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 32, i32 33>
+; NEON-NEXT:       [[TMP5:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 8, i32 9>
+; NEON-NEXT:       call void @llvm.arm.neon.vst4.p0i8.v2i32(i8* [[TMP1]], <2 x i32> [[TMP2]], <2 x i32> [[TMP3]], <2 x i32> [[TMP4]], <2 x i32> [[TMP5]], i32 4)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_general_mask_factor4_undefend(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <8 x i32> <i32 4, i32 16, i32 32, i32 8, i32 5, i32 17, i32 33, i32 undef>
+  store <8 x i32> %interleaved.vec, <8 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor4_undefmid(<8 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor4_undefmid(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <8 x i32>* %ptr to i8*
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 4, i32 5>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 16, i32 17>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 32, i32 33>
+; NEON-NEXT:       [[TMP5:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 8, i32 9>
+; NEON-NEXT:       call void @llvm.arm.neon.vst4.p0i8.v2i32(i8* [[TMP1]], <2 x i32> [[TMP2]], <2 x i32> [[TMP3]], <2 x i32> [[TMP4]], <2 x i32> [[TMP5]], i32 4)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_general_mask_factor4_undefmid(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <8 x i32> <i32 4, i32 undef, i32 32, i32 8, i32 5, i32 17, i32 undef, i32 9>
+  store <8 x i32> %interleaved.vec, <8 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor4_undefmulti(<8 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor4_undefmulti(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <8 x i32>* %ptr to i8*
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 4, i32 5>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 0, i32 1>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 0, i32 1>
+; NEON-NEXT:       [[TMP5:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <2 x i32> <i32 8, i32 9>
+; NEON-NEXT:       call void @llvm.arm.neon.vst4.p0i8.v2i32(i8* [[TMP1]], <2 x i32> [[TMP2]], <2 x i32> [[TMP3]], <2 x i32> [[TMP4]], <2 x i32> [[TMP5]], i32 4)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_general_mask_factor4_undefmulti(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <8 x i32> <i32 4, i32 undef, i32 undef, i32 8, i32 undef, i32 undef, i32 undef, i32 9>
+  store <8 x i32> %interleaved.vec, <8 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor3(<12 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor3(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <12 x i32>* %ptr to i8*
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 32, i32 33, i32 34, i32 35>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 16, i32 17, i32 18, i32 19>
+; NEON-NEXT:       call void @llvm.arm.neon.vst3.p0i8.v4i32(i8* [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> [[TMP3]], <4 x i32> [[TMP4]], i32 4)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_general_mask_factor3(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <12 x i32> <i32 4, i32 32, i32 16, i32 5, i32 33, i32 17, i32 6, i32 34, i32 18, i32 7, i32 35, i32 19>
+  store <12 x i32> %interleaved.vec, <12 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor3_undefmultimid(<12 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor3_undefmultimid(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <12 x i32>* %ptr to i8*
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 32, i32 33, i32 34, i32 35>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 16, i32 17, i32 18, i32 19>
+; NEON-NEXT:       call void @llvm.arm.neon.vst3.p0i8.v4i32(i8* [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> [[TMP3]], <4 x i32> [[TMP4]], i32 4)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_general_mask_factor3_undefmultimid(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <12 x i32> <i32 4, i32 32, i32 16, i32 undef, i32 33, i32 17, i32 undef, i32 34, i32 18, i32 7, i32 35, i32 19>
+  store <12 x i32> %interleaved.vec, <12 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor3_undef_fail(<12 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor3_undef_fail(
+; NEON-NOT:        @llvm.arm.neon
+; NEON:            ret void
+; NO_NEON-LABEL: @store_general_mask_factor3_undef_fail(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <12 x i32> <i32 4, i32 32, i32 16, i32 undef, i32 33, i32 17, i32 undef, i32 34, i32 18, i32 8, i32 35, i32 19>
+  store <12 x i32> %interleaved.vec, <12 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor3_undeflane(<12 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor3_undeflane(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <12 x i32>* %ptr to i8*
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 32, i32 33, i32 34, i32 35>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 16, i32 17, i32 18, i32 19>
+; NEON-NEXT:       call void @llvm.arm.neon.vst3.p0i8.v4i32(i8* [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> [[TMP3]], <4 x i32> [[TMP4]], i32 4)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_general_mask_factor3_undeflane(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <12 x i32> <i32 undef, i32 32, i32 16, i32 undef, i32 33, i32 17, i32 undef, i32 34, i32 18, i32 undef, i32 35, i32 19>
+  store <12 x i32> %interleaved.vec, <12 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor3_endstart_fail(<12 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor3_endstart_fail(
+; NEON-NOT:        @llvm.arm.neon
+; NEON:            ret void
+; NO_NEON-LABEL: @store_general_mask_factor3_endstart_fail(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <12 x i32> <i32 undef, i32 32, i32 16, i32 undef, i32 33, i32 17, i32 undef, i32 34, i32 18, i32 2, i32 35, i32 19>
+  store <12 x i32> %interleaved.vec, <12 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor3_endstart_pass(<12 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor3_endstart_pass(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <12 x i32>* %ptr to i8*
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 32, i32 33, i32 34, i32 35>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 16, i32 17, i32 18, i32 19>
+; NEON-NEXT:       call void @llvm.arm.neon.vst3.p0i8.v4i32(i8* [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> [[TMP3]], <4 x i32> [[TMP4]], i32 4)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_general_mask_factor3_endstart_pass(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <12 x i32> <i32 undef, i32 32, i32 16, i32 undef, i32 33, i32 17, i32 undef, i32 34, i32 18, i32 7, i32 35, i32 19>
+  store <12 x i32> %interleaved.vec, <12 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor3_midstart_fail(<12 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor3_midstart_fail(
+; NEON-NOT:        @llvm.arm.neon
+; NEON:            ret void
+; NO_NEON-LABEL: @store_general_mask_factor3_midstart_fail(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <12 x i32> <i32 undef, i32 32, i32 16, i32 0, i32 33, i32 17, i32 undef, i32 34, i32 18, i32 undef, i32 35, i32 19>
+  store <12 x i32> %interleaved.vec, <12 x i32>* %ptr, align 4
+  ret void
+}
+
+define void @store_general_mask_factor3_midstart_pass(<12 x i32>* %ptr, <32 x i32> %v0, <32 x i32> %v1) {
+; NEON-LABEL:    @store_general_mask_factor3_midstart_pass(
+; NEON-NEXT:       [[TMP1:%.*]] = bitcast <12 x i32>* %ptr to i8*
+; NEON-NEXT:       [[TMP2:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; NEON-NEXT:       [[TMP3:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 32, i32 33, i32 34, i32 35>
+; NEON-NEXT:       [[TMP4:%.*]] = shufflevector <32 x i32> %v0, <32 x i32> %v1, <4 x i32> <i32 16, i32 17, i32 18, i32 19>
+; NEON-NEXT:       call void @llvm.arm.neon.vst3.p0i8.v4i32(i8* [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> [[TMP3]], <4 x i32> [[TMP4]], i32 4)
+; NEON-NEXT:       ret void
+; NO_NEON-LABEL: @store_general_mask_factor3_midstart_pass(
+; NO_NEON-NOT:     @llvm.arm.neon
+; NO_NEON:         ret void
+;
+  %interleaved.vec = shufflevector <32 x i32> %v0, <32 x i32> %v1, <12 x i32> <i32 undef, i32 32, i32 16, i32 1, i32 33, i32 17, i32 undef, i32 34, i32 18, i32 undef, i32 35, i32 19>
+  store <12 x i32> %interleaved.vec, <12 x i32>* %ptr, align 4
+  ret void
+}
diff --git a/llvm/test/Transforms/InterleavedAccess/ARM/lit.local.cfg b/llvm/test/Transforms/InterleavedAccess/ARM/lit.local.cfg
new file mode 100644
index 00000000000..20e19aeb06f
--- /dev/null
+++ b/llvm/test/Transforms/InterleavedAccess/ARM/lit.local.cfg
@@ -0,0 +1,2 @@
+if not 'ARM' in config.root.targets:
+  config.unsupported = True