Implement sandboxing of NEON store instructions

BUG=http://code.google.com/p/nativeclient/issues/detail?id=3124

Review URL: https://codereview.chromium.org/11361249

6 files changed, 511 insertions, 3 deletions

diff --git a/test/NaCl/ARM/neon-vst1-sandboxing.ll b/test/NaCl/ARM/neon-vst1-sandboxing.ll
new file mode 100644
index 0000000000..8fd580bb49
--- /dev/null
+++ b/test/NaCl/ARM/neon-vst1-sandboxing.ll
@@ -0,0 +1,116 @@
+; RUN: llc -mtriple=armv7-unknown-nacl -mattr=+neon -sfi-store -filetype=obj %s -o - \
+; RUN:  | llvm-objdump -disassemble -triple armv7 - | FileCheck %s
+
+define void @vst1i8(i8* %A, <8 x i8>* %B) nounwind {
+  %tmp1 = load <8 x i8>* %B
+  call void @llvm.arm.neon.vst1.v8i8(i8* %A, <8 x i8> %tmp1, i32 16)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst1.8 {{{d[0-9]+}}}, [r0, :64]
+  ret void
+}
+
+define void @vst1i16(i16* %A, <4 x i16>* %B) nounwind {
+  %tmp0 = bitcast i16* %A to i8*
+  %tmp1 = load <4 x i16>* %B
+  call void @llvm.arm.neon.vst1.v4i16(i8* %tmp0, <4 x i16> %tmp1, i32 1)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst1.16 {{{d[0-9]+}}}, [r0]
+  ret void
+}
+
+define void @vst1i32(i32* %A, <2 x i32>* %B) nounwind {
+  %tmp0 = bitcast i32* %A to i8*
+  %tmp1 = load <2 x i32>* %B
+  call void @llvm.arm.neon.vst1.v2i32(i8* %tmp0, <2 x i32> %tmp1, i32 1)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst1.32 {{{d[0-9]+}}}, [r0]
+  ret void
+}
+
+define void @vst1f(float* %A, <2 x float>* %B) nounwind {
+  %tmp0 = bitcast float* %A to i8*
+  %tmp1 = load <2 x float>* %B
+  call void @llvm.arm.neon.vst1.v2f32(i8* %tmp0, <2 x float> %tmp1, i32 1)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst1.32 {{{d[0-9]+}}}, [r0]
+  ret void
+}
+
+define void @vst1i64(i64* %A, <1 x i64>* %B) nounwind {
+  %tmp0 = bitcast i64* %A to i8*
+  %tmp1 = load <1 x i64>* %B
+  call void @llvm.arm.neon.vst1.v1i64(i8* %tmp0, <1 x i64> %tmp1, i32 1)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst1.64 {{{d[0-9]+}}}, [r0]
+  ret void
+}
+
+define void @vst1Qi8(i8* %A, <16 x i8>* %B) nounwind {
+  %tmp1 = load <16 x i8>* %B
+  call void @llvm.arm.neon.vst1.v16i8(i8* %A, <16 x i8> %tmp1, i32 8)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst1.8 {{{d[0-9]+, d[0-9]+}}}, [r0, :64]
+  ret void
+}
+
+define void @vst1Qi16(i16* %A, <8 x i16>* %B) nounwind {
+  %tmp0 = bitcast i16* %A to i8*
+  %tmp1 = load <8 x i16>* %B
+  call void @llvm.arm.neon.vst1.v8i16(i8* %tmp0, <8 x i16> %tmp1, i32 32)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst1.16 {{{d[0-9]+, d[0-9]+}}}, [r0, :128]
+  ret void
+}
+
+define void @vst1Qi32(i32* %A, <4 x i32>* %B) nounwind {
+  %tmp0 = bitcast i32* %A to i8*
+  %tmp1 = load <4 x i32>* %B
+  call void @llvm.arm.neon.vst1.v4i32(i8* %tmp0, <4 x i32> %tmp1, i32 1)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst1.32 {{{d[0-9]+, d[0-9]+}}}, [r0]
+  ret void
+}
+
+define void @vst1Qf(float* %A, <4 x float>* %B) nounwind {
+  %tmp0 = bitcast float* %A to i8*
+  %tmp1 = load <4 x float>* %B
+  call void @llvm.arm.neon.vst1.v4f32(i8* %tmp0, <4 x float> %tmp1, i32 1)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst1.32 {{{d[0-9]+, d[0-9]+}}}, [r0]
+  ret void
+}
+
+define void @vst1Qi64(i64* %A, <2 x i64>* %B) nounwind {
+  %tmp0 = bitcast i64* %A to i8*
+  %tmp1 = load <2 x i64>* %B
+  call void @llvm.arm.neon.vst1.v2i64(i8* %tmp0, <2 x i64> %tmp1, i32 1)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst1.64 {{{d[0-9]+, d[0-9]+}}}, [r0]
+  ret void
+}
+
+;Check for a post-increment updating store.
+define void @vst1f_update(float** %ptr, <2 x float>* %B) nounwind {
+  %A = load float** %ptr
+  %tmp0 = bitcast float* %A to i8*
+  %tmp1 = load <2 x float>* %B
+  call void @llvm.arm.neon.vst1.v2f32(i8* %tmp0, <2 x float> %tmp1, i32 1)
+; CHECK:         bic r1, r1, #3221225472
+; CHECK-NEXT:    vst1.32 {{{d[0-9]+}}}, [r1]!
+  %tmp2 = getelementptr float* %A, i32 2
+  store float* %tmp2, float** %ptr
+  ret void
+}
+
+declare void @llvm.arm.neon.vst1.v8i8(i8*, <8 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst1.v4i16(i8*, <4 x i16>, i32) nounwind
+declare void @llvm.arm.neon.vst1.v2i32(i8*, <2 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst1.v2f32(i8*, <2 x float>, i32) nounwind
+declare void @llvm.arm.neon.vst1.v1i64(i8*, <1 x i64>, i32) nounwind
+
+declare void @llvm.arm.neon.vst1.v16i8(i8*, <16 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst1.v8i16(i8*, <8 x i16>, i32) nounwind
+declare void @llvm.arm.neon.vst1.v4i32(i8*, <4 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst1.v4f32(i8*, <4 x float>, i32) nounwind
+declare void @llvm.arm.neon.vst1.v2i64(i8*, <2 x i64>, i32) nounwind
+
diff --git a/test/NaCl/ARM/neon-vst2-sandboxing.ll b/test/NaCl/ARM/neon-vst2-sandboxing.ll
new file mode 100644
index 0000000000..e87373c174
--- /dev/null
+++ b/test/NaCl/ARM/neon-vst2-sandboxing.ll
@@ -0,0 +1,95 @@
+; RUN: llc -mtriple=armv7-unknown-nacl -mattr=+neon -sfi-store -filetype=obj %s -o - \
+; RUN:  | llvm-objdump -disassemble -triple armv7 - | FileCheck %s
+
+define void @vst2i8(i8* %A, <8 x i8>* %B) nounwind {
+  %tmp1 = load <8 x i8>* %B
+  call void @llvm.arm.neon.vst2.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 8)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst2.8 {{{d[0-9]+, d[0-9]+}}}, [r0, :64]
+  ret void
+}
+
+define void @vst2i16(i16* %A, <4 x i16>* %B) nounwind {
+  %tmp0 = bitcast i16* %A to i8*
+  %tmp1 = load <4 x i16>* %B
+  call void @llvm.arm.neon.vst2.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 32)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst2.16 {{{d[0-9]+, d[0-9]+}}}, [r0, :128]
+  ret void
+}
+
+define void @vst2i32(i32* %A, <2 x i32>* %B) nounwind {
+  %tmp0 = bitcast i32* %A to i8*
+  %tmp1 = load <2 x i32>* %B
+  call void @llvm.arm.neon.vst2.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst2.32 {{{d[0-9]+, d[0-9]+}}}, [r0]
+  ret void
+}
+
+define void @vst2f(float* %A, <2 x float>* %B) nounwind {
+  %tmp0 = bitcast float* %A to i8*
+  %tmp1 = load <2 x float>* %B
+  call void @llvm.arm.neon.vst2.v2f32(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, i32 1)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst2.32 {{{d[0-9]+, d[0-9]+}}}, [r0]
+  ret void
+}
+
+define void @vst2Qi8(i8* %A, <16 x i8>* %B) nounwind {
+  %tmp1 = load <16 x i8>* %B
+  call void @llvm.arm.neon.vst2.v16i8(i8* %A, <16 x i8> %tmp1, <16 x i8> %tmp1, i32 8)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst2.8 {{{d[0-9]+, d[0-9]+, d[0-9]+, d[0-9]+}}}, [r0, :64]
+  ret void
+}
+
+define void @vst2Qi16(i16* %A, <8 x i16>* %B) nounwind {
+  %tmp0 = bitcast i16* %A to i8*
+  %tmp1 = load <8 x i16>* %B
+  call void @llvm.arm.neon.vst2.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 16)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst2.16 {{{d[0-9]+, d[0-9]+, d[0-9]+, d[0-9]+}}}, [r0, :128]
+  ret void
+}
+
+define void @vst2Qi32(i32* %A, <4 x i32>* %B) nounwind {
+  %tmp0 = bitcast i32* %A to i8*
+  %tmp1 = load <4 x i32>* %B
+  call void @llvm.arm.neon.vst2.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 64)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst2.32 {{{d[0-9]+, d[0-9]+, d[0-9]+, d[0-9]+}}}, [r0, :256]
+  ret void
+}
+
+define void @vst2Qf(float* %A, <4 x float>* %B) nounwind {
+  %tmp0 = bitcast float* %A to i8*
+  %tmp1 = load <4 x float>* %B
+  call void @llvm.arm.neon.vst2.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, i32 1)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst2.32 {{{d[0-9]+, d[0-9]+, d[0-9]+, d[0-9]+}}}, [r0]
+  ret void
+}
+
+;Check for a post-increment updating store with register increment.
+define void @vst2i8_update(i8** %ptr, <8 x i8>* %B, i32 %inc) nounwind {
+  %A = load i8** %ptr
+  %tmp1 = load <8 x i8>* %B
+  call void @llvm.arm.neon.vst2.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 4)
+; CHECK:         bic r1, r1, #3221225472
+; CHECK-NEXT:    vst2.8 {{{d[0-9]+, d[0-9]+}}}, [r1], r2
+  %tmp2 = getelementptr i8* %A, i32 %inc
+  store i8* %tmp2, i8** %ptr
+  ret void
+}
+
+declare void @llvm.arm.neon.vst2.v8i8(i8*, <8 x i8>, <8 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst2.v4i16(i8*, <4 x i16>, <4 x i16>, i32) nounwind
+declare void @llvm.arm.neon.vst2.v2i32(i8*, <2 x i32>, <2 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst2.v2f32(i8*, <2 x float>, <2 x float>, i32) nounwind
+declare void @llvm.arm.neon.vst2.v1i64(i8*, <1 x i64>, <1 x i64>, i32) nounwind
+
+declare void @llvm.arm.neon.vst2.v16i8(i8*, <16 x i8>, <16 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst2.v8i16(i8*, <8 x i16>, <8 x i16>, i32) nounwind
+declare void @llvm.arm.neon.vst2.v4i32(i8*, <4 x i32>, <4 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst2.v4f32(i8*, <4 x float>, <4 x float>, i32) nounwind
diff --git a/test/NaCl/ARM/neon-vst3-sandboxing.ll b/test/NaCl/ARM/neon-vst3-sandboxing.ll
new file mode 100644
index 0000000000..b496c0c592
--- /dev/null
+++ b/test/NaCl/ARM/neon-vst3-sandboxing.ll
@@ -0,0 +1,48 @@
+; RUN: llc -mtriple=armv7-unknown-nacl -mattr=+neon -sfi-store -filetype=obj %s -o - \
+; RUN:  | llvm-objdump -disassemble -triple armv7 - | FileCheck %s
+
+define void @vst3i8(i8* %A, <8 x i8>* %B) nounwind {
+  %tmp1 = load <8 x i8>* %B
+  call void @llvm.arm.neon.vst3.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 32)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst3.8 {d{{[0-9]+}}, d{{[0-9]+}}, d{{[0-9]+}}}, [r0, :64]
+  ret void
+}
+
+define void @vst3i16(i16* %A, <4 x i16>* %B) nounwind {
+  %tmp0 = bitcast i16* %A to i8*
+  %tmp1 = load <4 x i16>* %B
+  call void @llvm.arm.neon.vst3.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst3.16 {d{{[0-9]+}}, d{{[0-9]+}}, d{{[0-9]+}}}, [r0]
+  ret void
+}
+
+define void @vst3i32(i32* %A, <2 x i32>* %B) nounwind {
+  %tmp0 = bitcast i32* %A to i8*
+  %tmp1 = load <2 x i32>* %B
+  call void @llvm.arm.neon.vst3.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst3.32 {d{{[0-9]+}}, d{{[0-9]+}}, d{{[0-9]+}}}, [r0]
+  ret void
+}
+
+;Check for a post-increment updating store.
+define void @vst3Qi16_update(i16** %ptr, <8 x i16>* %B) nounwind {
+  %A = load i16** %ptr
+  %tmp0 = bitcast i16* %A to i8*
+  %tmp1 = load <8 x i16>* %B
+  call void @llvm.arm.neon.vst3.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 1)
+; CHECK:         bic r1, r1, #3221225472
+; CHECK-NEXT:    vst3.16 {d{{[0-9]+}}, d{{[0-9]+}}, d{{[0-9]+}}}, [r1]!
+  %tmp2 = getelementptr i16* %A, i32 24
+  store i16* %tmp2, i16** %ptr
+  ret void
+}
+
+declare void @llvm.arm.neon.vst3.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst3.v4i16(i8*, <4 x i16>, <4 x i16>, <4 x i16>, i32) nounwind
+declare void @llvm.arm.neon.vst3.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst3.v2f32(i8*, <2 x float>, <2 x float>, <2 x float>, i32) nounwind
+
+declare void @llvm.arm.neon.vst3.v8i16(i8*, <8 x i16>, <8 x i16>, <8 x i16>, i32) nounwind
diff --git a/test/NaCl/ARM/neon-vst4-sandboxing.ll b/test/NaCl/ARM/neon-vst4-sandboxing.ll
new file mode 100644
index 0000000000..032f194231
--- /dev/null
+++ b/test/NaCl/ARM/neon-vst4-sandboxing.ll
@@ -0,0 +1,53 @@
+; RUN: llc -mtriple=armv7-unknown-nacl -mattr=+neon -sfi-store -filetype=obj %s -o - \
+; RUN:  | llvm-objdump -disassemble -triple armv7 - | FileCheck %s
+
+define void @vst4i8(i8* %A, <8 x i8>* %B) nounwind {
+  %tmp1 = load <8 x i8>* %B
+  call void @llvm.arm.neon.vst4.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 8)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst4.8 {d{{[0-9]+}}, d{{[0-9]+}}, d{{[0-9]+}}, d{{[0-9]+}}}, [r0, :64]
+  ret void
+}
+
+define void @vst4i16(i16* %A, <4 x i16>* %B) nounwind {
+  %tmp0 = bitcast i16* %A to i8*
+  %tmp1 = load <4 x i16>* %B
+  call void @llvm.arm.neon.vst4.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 16)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst4.16 {d{{[0-9]+}}, d{{[0-9]+}}, d{{[0-9]+}}, d{{[0-9]+}}}, [r0, :128]
+  ret void
+}
+
+define void @vst4i32(i32* %A, <2 x i32>* %B) nounwind {
+  %tmp0 = bitcast i32* %A to i8*
+  %tmp1 = load <2 x i32>* %B
+  call void @llvm.arm.neon.vst4.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 32)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst4.32 {d{{[0-9]+}}, d{{[0-9]+}}, d{{[0-9]+}}, d{{[0-9]+}}}, [r0, :256]
+  ret void
+}
+
+;Check for a post-increment updating store.
+define void @vst4Qf_update(float** %ptr, <4 x float>* %B) nounwind {
+  %A = load float** %ptr
+  %tmp0 = bitcast float* %A to i8*
+  %tmp1 = load <4 x float>* %B
+  call void @llvm.arm.neon.vst4.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, i32 1)
+; CHECK:         bic r1, r1, #3221225472
+; CHECK-NEXT:    vst4.32 {d{{[0-9]+}}, d{{[0-9]+}}, d{{[0-9]+}}, d{{[0-9]+}}}, [r1]!
+  %tmp2 = getelementptr float* %A, i32 16
+  store float* %tmp2, float** %ptr
+  ret void
+}
+
+declare void @llvm.arm.neon.vst4.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst4.v4i16(i8*, <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16>, i32) nounwind
+declare void @llvm.arm.neon.vst4.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst4.v2f32(i8*, <2 x float>, <2 x float>, <2 x float>, <2 x float>, i32) nounwind
+declare void @llvm.arm.neon.vst4.v1i64(i8*, <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64>, i32) nounwind
+
+declare void @llvm.arm.neon.vst4.v16i8(i8*, <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst4.v8i16(i8*, <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16>, i32) nounwind
+declare void @llvm.arm.neon.vst4.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst4.v4f32(i8*, <4 x float>, <4 x float>, <4 x float>, <4 x float>, i32) nounwind
+
diff --git a/test/NaCl/ARM/neon-vstlane-sandboxing.ll b/test/NaCl/ARM/neon-vstlane-sandboxing.ll
new file mode 100644
index 0000000000..5b4dc63a14
--- /dev/null
+++ b/test/NaCl/ARM/neon-vstlane-sandboxing.ll
@@ -0,0 +1,196 @@
+; RUN: llc -mtriple=armv7-unknown-nacl -mattr=+neon -sfi-store -filetype=obj %s -o - \
+; RUN:  | llvm-objdump -disassemble -triple armv7 - | FileCheck %s
+
+define void @vst1lanei8(i8* %A, <8 x i8>* %B) nounwind {
+  %tmp1 = load <8 x i8>* %B
+        %tmp2 = extractelement <8 x i8> %tmp1, i32 3
+        store i8 %tmp2, i8* %A, align 8
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst1.8 {d{{[0-9]+}}[3]}, [r0]
+  ret void
+}
+
+define void @vst1lanei16(i16* %A, <4 x i16>* %B) nounwind {
+  %tmp1 = load <4 x i16>* %B
+        %tmp2 = extractelement <4 x i16> %tmp1, i32 2
+        store i16 %tmp2, i16* %A, align 8
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst1.16 {d{{[0-9]+}}[2]}, [r0, :16]
+  ret void
+}
+
+define void @vst1lanei32(i32* %A, <2 x i32>* %B) nounwind {
+  %tmp1 = load <2 x i32>* %B
+        %tmp2 = extractelement <2 x i32> %tmp1, i32 1
+        store i32 %tmp2, i32* %A, align 8
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst1.32 {d{{[0-9]+}}[1]}, [r0, :32]
+  ret void
+}
+
+define void @vst1laneQi8(i8* %A, <16 x i8>* %B) nounwind {
+  %tmp1 = load <16 x i8>* %B
+        %tmp2 = extractelement <16 x i8> %tmp1, i32 9
+        store i8 %tmp2, i8* %A, align 8
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst1.8 {d{{[0-9]+}}[1]}, [r0]
+  ret void
+}
+
+define void @vst1laneQi16(i16* %A, <8 x i16>* %B) nounwind {
+  %tmp1 = load <8 x i16>* %B
+        %tmp2 = extractelement <8 x i16> %tmp1, i32 5
+        store i16 %tmp2, i16* %A, align 8
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst1.16 {d{{[0-9]+}}[1]}, [r0, :16]
+  ret void
+}
+
+define void @vst2lanei8(i8* %A, <8 x i8>* %B) nounwind {
+  %tmp1 = load <8 x i8>* %B
+  call void @llvm.arm.neon.vst2lane.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 4)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst2.8 {d{{[0-9]+}}[1], d{{[0-9]+}}[1]}, [r0, :16]
+  ret void
+}
+
+define void @vst2lanei16(i16* %A, <4 x i16>* %B) nounwind {
+  %tmp0 = bitcast i16* %A to i8*
+  %tmp1 = load <4 x i16>* %B
+  call void @llvm.arm.neon.vst2lane.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1, i32 8)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst2.16 {d{{[0-9]+}}[1], d{{[0-9]+}}[1]}, [r0, :32]
+  ret void
+}
+
+define void @vst2lanei32(i32* %A, <2 x i32>* %B) nounwind {
+  %tmp0 = bitcast i32* %A to i8*
+  %tmp1 = load <2 x i32>* %B
+  call void @llvm.arm.neon.vst2lane.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1, i32 1)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst2.32 {d{{[0-9]+}}[1], d{{[0-9]+}}[1]}, [r0]
+  ret void
+}
+
+define void @vst2laneQi16(i16* %A, <8 x i16>* %B) nounwind {
+  %tmp0 = bitcast i16* %A to i8*
+  %tmp1 = load <8 x i16>* %B
+  call void @llvm.arm.neon.vst2lane.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 5, i32 1)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst2.16 {d{{[0-9]+}}[1], d{{[0-9]+}}[1]}, [r0]
+  ret void
+}
+
+define void @vst2laneQi32(i32* %A, <4 x i32>* %B) nounwind {
+  %tmp0 = bitcast i32* %A to i8*
+  %tmp1 = load <4 x i32>* %B
+  call void @llvm.arm.neon.vst2lane.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 2, i32 16)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst2.32 {d{{[0-9]+}}[0], d{{[0-9]+}}[0]}, [r0, :64]
+  ret void
+}
+
+define void @vst3lanei8(i8* %A, <8 x i8>* %B) nounwind {
+  %tmp1 = load <8 x i8>* %B
+  call void @llvm.arm.neon.vst3lane.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 1)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst3.8 {d{{[0-9]+}}[1], d{{[0-9]+}}[1], d{{[0-9]+}}[1]}, [r0]
+  ret void
+}
+
+define void @vst3lanei16(i16* %A, <4 x i16>* %B) nounwind {
+  %tmp0 = bitcast i16* %A to i8*
+  %tmp1 = load <4 x i16>* %B
+  call void @llvm.arm.neon.vst3lane.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1, i32 8)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst3.16 {d{{[0-9]+}}[1], d{{[0-9]+}}[1], d{{[0-9]+}}[1]}, [r0]
+  ret void
+}
+
+define void @vst3lanei32(i32* %A, <2 x i32>* %B) nounwind {
+  %tmp0 = bitcast i32* %A to i8*
+  %tmp1 = load <2 x i32>* %B
+  call void @llvm.arm.neon.vst3lane.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1, i32 1)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst3.32 {d{{[0-9]+}}[1], d{{[0-9]+}}[1], d{{[0-9]+}}[1]}, [r0]
+  ret void
+}
+
+define void @vst4lanei8(i8* %A, <8 x i8>* %B) nounwind {
+  %tmp1 = load <8 x i8>* %B
+  call void @llvm.arm.neon.vst4lane.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 8)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst4.8 {d{{[0-9]+}}[1], d{{[0-9]+}}[1], d{{[0-9]+}}[1], d{{[0-9]+}}[1]}, [r0, :32]
+  ret void
+}
+
+define void @vst4lanei16(i16* %A, <4 x i16>* %B) nounwind {
+  %tmp0 = bitcast i16* %A to i8*
+  %tmp1 = load <4 x i16>* %B
+  call void @llvm.arm.neon.vst4lane.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1, i32 1)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst4.16 {d{{[0-9]+}}[1], d{{[0-9]+}}[1], d{{[0-9]+}}[1], d{{[0-9]+}}[1]}, [r0]
+  ret void
+}
+
+define void @vst4lanei32(i32* %A, <2 x i32>* %B) nounwind {
+  %tmp0 = bitcast i32* %A to i8*
+  %tmp1 = load <2 x i32>* %B
+  call void @llvm.arm.neon.vst4lane.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1, i32 16)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst4.32 {d{{[0-9]+}}[1], d{{[0-9]+}}[1], d{{[0-9]+}}[1], d{{[0-9]+}}[1]}, [r0, :128]
+  ret void
+}
+
+define void @vst4laneQi16(i16* %A, <8 x i16>* %B) nounwind {
+  %tmp0 = bitcast i16* %A to i8*
+  %tmp1 = load <8 x i16>* %B
+  call void @llvm.arm.neon.vst4lane.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 7, i32 16)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst4.16 {d{{[0-9]+}}[3], d{{[0-9]+}}[3], d{{[0-9]+}}[3], d{{[0-9]+}}[3]}, [r0, :64]
+  ret void
+}
+
+define void @vst4laneQi32(i32* %A, <4 x i32>* %B) nounwind {
+  %tmp0 = bitcast i32* %A to i8*
+  %tmp1 = load <4 x i32>* %B
+  call void @llvm.arm.neon.vst4lane.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 2, i32 1)
+; CHECK:         bic r0, r0, #3221225472
+; CHECK-NEXT:    vst4.32 {d{{[0-9]+}}[0], d{{[0-9]+}}[0], d{{[0-9]+}}[0], d{{[0-9]+}}[0]}, [r0]
+  ret void
+}
+
+declare void @llvm.arm.neon.vst2lane.v8i8(i8*, <8 x i8>, <8 x i8>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst2lane.v4i16(i8*, <4 x i16>, <4 x i16>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst2lane.v2i32(i8*, <2 x i32>, <2 x i32>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst2lane.v2f32(i8*, <2 x float>, <2 x float>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst2lane.v8i16(i8*, <8 x i16>, <8 x i16>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst2lane.v4i32(i8*, <4 x i32>, <4 x i32>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst2lane.v4f32(i8*, <4 x float>, <4 x float>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst3lane.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst3lane.v4i16(i8*, <4 x i16>, <4 x i16>, <4 x i16>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst3lane.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst3lane.v2f32(i8*, <2 x float>, <2 x float>, <2 x float>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst3lane.v8i16(i8*, <8 x i16>, <8 x i16>, <8 x i16>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst3lane.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst3lane.v4f32(i8*, <4 x float>, <4 x float>, <4 x float>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst4lane.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst4lane.v4i16(i8*, <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst4lane.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst4lane.v2f32(i8*, <2 x float>, <2 x float>, <2 x float>, <2 x float>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst4lane.v8i16(i8*, <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst4lane.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst4lane.v4f32(i8*, <4 x float>, <4 x float>, <4 x float>, <4 x float>, i32, i32) nounwind
+
+;Check for a post-increment updating store with register increment.
+define void @vst2lanei16_update(i16** %ptr, <4 x i16>* %B, i32 %inc) nounwind {
+  %A = load i16** %ptr
+  %tmp0 = bitcast i16* %A to i8*
+  %tmp1 = load <4 x i16>* %B
+  call void @llvm.arm.neon.vst2lane.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1, i32 2)
+; CHECK:         bic r1, r1, #3221225472
+; CHECK-NEXT:    vst2.16 {d{{[0-9]+}}[1], d{{[0-9]+}}[1]}, [r1], r2
+  %tmp2 = getelementptr i16* %A, i32 %inc
+  store i16* %tmp2, i16** %ptr
+  ret void
+}
diff --git a/test/NaCl/ARM/vstr-sandboxing1.ll b/test/NaCl/ARM/vstr-sandboxing1.ll
index 06aeff1037..6646cbc717 100644
--- a/test/NaCl/ARM/vstr-sandboxing1.ll
+++ b/test/NaCl/ARM/vstr-sandboxing1.ll
@@ -5,9 +5,9 @@ define void @test_vstr_sandbox(<8 x i8>* %ptr) nounwind {
   %1 = insertelement <8 x i8> undef, i8 -128, i32 0
   %2 = shufflevector <8 x i8> %1, <8 x i8> undef, <8 x i32> zeroinitializer
   store <8 x i8> %2, <8 x i8>* %ptr, align 8
-  ret void
-}
-
 ; CHECK:         bic r0, r0, #3221225472
 ; CHECK-NEXT:    vstr {{[0-9a-z]+}}, [r0]
 
+  ret void
+}
+