Fill out support for Thumb2 encodings of NEON instructions.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118854 91177308-0d34-0410-b5e6-96231b3b80d8

12 files changed, 553 insertions, 0 deletions

diff --git a/lib/Target/ARM/ARMCodeEmitter.cpp b/lib/Target/ARM/ARMCodeEmitter.cpp
index cd4ac6a..b068c64 100644
--- a/lib/Target/ARM/ARMCodeEmitter.cpp
+++ b/lib/Target/ARM/ARMCodeEmitter.cpp
@@ -165,6 +165,8 @@ namespace {
       const { return 0; }
     unsigned NEONThumb2LoadStorePostEncoder(const MachineInstr &MI,unsigned Val) 
       const { return 0; }
+    unsigned NEONThumb2DupPostEncoder(const MachineInstr &MI,unsigned Val) 
+      const { return 0; }
     unsigned getBranchTargetOpValue(const MachineInstr &MI, unsigned Op)
       const { return 0; }
     unsigned getCCOutOpValue(const MachineInstr &MI, unsigned Op)
diff --git a/lib/Target/ARM/ARMInstrFormats.td b/lib/Target/ARM/ARMInstrFormats.td
index 56aa4b0..251e5c2 100644
--- a/lib/Target/ARM/ARMInstrFormats.td
+++ b/lib/Target/ARM/ARMInstrFormats.td
@@ -2046,6 +2046,8 @@ class NVLaneOp<bits<8> opcod1, bits<4> opcod2, bits<2> opcod3,
   let Pattern = pattern;
   list<Predicate> Predicates = [HasNEON];
   
+  string PostEncoderMethod = "NEONThumb2DupPostEncoder";
+  
   bits<5> V;
   bits<4> R;
   bits<4> p;
diff --git a/lib/Target/ARM/ARMMCCodeEmitter.cpp b/lib/Target/ARM/ARMMCCodeEmitter.cpp
index a407aca..ebd0d9b 100644
--- a/lib/Target/ARM/ARMMCCodeEmitter.cpp
+++ b/lib/Target/ARM/ARMMCCodeEmitter.cpp
@@ -176,6 +176,8 @@ public:
                                       unsigned EncodedValue) const;
   unsigned NEONThumb2LoadStorePostEncoder(const MCInst &MI,
                                       unsigned EncodedValue) const;
+  unsigned NEONThumb2DupPostEncoder(const MCInst &MI,
+                                      unsigned EncodedValue) const;
 
   void EmitByte(unsigned char C, raw_ostream &OS) const {
     OS << (char)C;
@@ -234,6 +236,21 @@ unsigned ARMMCCodeEmitter::NEONThumb2LoadStorePostEncoder(const MCInst &MI,
   return EncodedValue;
 }
 
+/// NEONThumb2DupPostEncoder - Post-process encoded NEON vdup
+/// instructions, and rewrite them to their Thumb2 form if we are currently in 
+/// Thumb2 mode.
+unsigned ARMMCCodeEmitter::NEONThumb2DupPostEncoder(const MCInst &MI,
+                                                 unsigned EncodedValue) const {
+  const ARMSubtarget &Subtarget = TM.getSubtarget<ARMSubtarget>();
+  if (Subtarget.isThumb2()) {
+    EncodedValue &= 0x00FFFFFF;
+    EncodedValue |= 0xEE000000;
+  }
+  
+  return EncodedValue;
+}
+
+
 
 /// getMachineOpValue - Return binary encoding of operand. If the machine
 /// operand requires relocation, record the relocation and return zero.
diff --git a/test/MC/ARM/neont2-absdiff-encoding.s b/test/MC/ARM/neont2-absdiff-encoding.s
new file mode 100644
index 0000000..be8e798
--- /dev/null
+++ b/test/MC/ARM/neont2-absdiff-encoding.s
@@ -0,0 +1,86 @@
+@ RUN: llvm-mc -mcpu=cortex-a8 -triple thumb-unknown-unkown -show-encoding < %s | FileCheck %s
+@ XFAIL: *
+@ NOTE: This currently fails because the ASM parser doesn't parse vabal.
+
+.code 16
+
+@ CHECK: vabd.s8	d16, d16, d17           @ encoding: [0xa1,0x07,0x40,0xef]
+	vabd.s8	d16, d16, d17
+@ CHECK: vabd.s16	d16, d16, d17   @ encoding: [0xa1,0x07,0x50,0xef]
+	vabd.s16	d16, d16, d17
+@ CHECK: vabd.s32	d16, d16, d17   @ encoding: [0xa1,0x07,0x60,0xef]
+	vabd.s32	d16, d16, d17
+@ CHECK: vabd.u8	d16, d16, d17           @ encoding: [0xa1,0x07,0x40,0xff]
+	vabd.u8	d16, d16, d17
+@ CHECK: vabd.u16	d16, d16, d17   @ encoding: [0xa1,0x07,0x50,0xff]
+	vabd.u16	d16, d16, d17
+  @ CHECK: vabd.u32	d16, d16, d17   @ encoding: [0xa1,0x07,0x60,0xff]
+	vabd.u32	d16, d16, d17
+@ CHECK: vabd.f32	d16, d16, d17   @ encoding: [0xa1,0x0d,0x60,0xff]
+	vabd.f32	d16, d16, d17
+@ CHECK: vabd.s8	q8, q8, q9              @ encoding: [0xe2,0x07,0x40,0xef]
+	vabd.s8	q8, q8, q9
+@ CHECK: vabd.s16	q8, q8, q9      @ encoding: [0xe2,0x07,0x50,0xef]
+	vabd.s16	q8, q8, q9
+@ CHECK: vabd.s32	q8, q8, q9      @ encoding: [0xe2,0x07,0x60,0xef]
+	vabd.s32	q8, q8, q9
+@ CHECK: vabd.u8	q8, q8, q9              @ encoding: [0xe2,0x07,0x40,0xff]
+	vabd.u8	q8, q8, q9
+@ CHECK: vabd.u16	q8, q8, q9      @ encoding: [0xe2,0x07,0x50,0xff]
+	vabd.u16	q8, q8, q9
+@ CHECK: vabd.u32	q8, q8, q9      @ encoding: [0xe2,0x07,0x60,0xff]
+	vabd.u32	q8, q8, q9
+@ CHECK: vabd.f32	q8, q8, q9      @ encoding: [0xe2,0x0d,0x60,0xff]
+	vabd.f32	q8, q8, q9
+
+@ CHECK: vabdl.s8	q8, d16, d17    @ encoding: [0xa1,0x07,0xc0,0xef]
+	vabdl.s8	q8, d16, d17
+@ CHECK: vabdl.s16	q8, d16, d17    @ encoding: [0xa1,0x07,0xd0,0xef]
+	vabdl.s16	q8, d16, d17
+@ CHECK: vabdl.s32	q8, d16, d17    @ encoding: [0xa1,0x07,0xe0,0xef]
+	vabdl.s32	q8, d16, d17
+@ CHECK: vabdl.u8	q8, d16, d17    @ encoding: [0xa1,0x07,0xc0,0xff]
+	vabdl.u8	q8, d16, d17
+@ CHECK: vabdl.u16	q8, d16, d17    @ encoding: [0xa1,0x07,0xd0,0xff]
+	vabdl.u16	q8, d16, d17
+@ CHECK: vabdl.u32	q8, d16, d17    @ encoding: [0xa1,0x07,0xe0,0xff]
+	vabdl.u32	q8, d16, d17
+
+@ CHECK: vaba.s8	d16, d18, d17           @ encoding: [0xb1,0x07,0x42,0xef]
+	vaba.s8	d16, d18, d17
+@ CHECK: vaba.s16	d16, d18, d17   @ encoding: [0xb1,0x07,0x52,0xef]
+	vaba.s16	d16, d18, d17
+@ CHECK: vaba.s32	d16, d18, d17   @ encoding: [0xb1,0x07,0x62,0xef]
+	vaba.s32	d16, d18, d17
+@ CHECK: vaba.u8	d16, d18, d17           @ encoding: [0xb1,0x07,0x42,0xff]
+	vaba.u8	d16, d18, d17
+@ CHECK: vaba.u16	d16, d18, d17   @ encoding: [0xb1,0x07,0x52,0xff]
+	vaba.u16	d16, d18, d17
+@ CHECK: vaba.u32	d16, d18, d17   @ encoding: [0xb1,0x07,0x62,0xff]
+	vaba.u32	d16, d18, d17
+@ CHECK: vaba.s8	q9, q8, q10             @ encoding: [0xf4,0x27,0x40,0xef]
+	vaba.s8	q9, q8, q10
+@ CHECK: vaba.s16	q9, q8, q10     @ encoding: [0xf4,0x27,0x50,0xef]
+	vaba.s16	q9, q8, q10
+@ CHECK: vaba.s32	q9, q8, q10     @ encoding: [0xf4,0x27,0x60,0xef]
+	vaba.s32	q9, q8, q10
+@ CHECK: vaba.u8	q9, q8, q10             @ encoding: [0xf4,0x27,0x40,0xff]
+	vaba.u8	q9, q8, q10
+@ CHECK: vaba.u16	q9, q8, q10     @ encoding: [0xf4,0x27,0x50,0xff]
+	vaba.u16	q9, q8, q10
+@ CHECK: vaba.u32	q9, q8, q10     @ encoding: [0xf4,0x27,0x60,0xff]
+	vaba.u32	q9, q8, q10
+
+@ CHECK: vabal.s8	q8, d19, d18    @ encoding: [0xa2,0x05,0xc3,0xef]
+	vabal.s8	q8, d19, d18
+@ CHECK: vabal.s16	q8, d19, d18    @ encoding: [0xa2,0x05,0xd3,0xef]
+	vabal.s16	q8, d19, d18
+@ CHECK: vabal.s32	q8, d19, d18    @ encoding: [0xa2,0x05,0xe3,0xef]
+	vabal.s32	q8, d19, d18
+@ CHECK: vabal.u8	q8, d19, d18    @ encoding: [0xa2,0x05,0xc3,0xff]
+	vabal.u8	q8, d19, d18
+@ CHECK: 	vabal.u16	q8, d19, d18    @ encoding: [0xa2,0x05,0xd3,0xff]
+	vabal.u16	q8, d19, d18
+@ CHECK: vabal.u32	q8, d19, d18    @ encoding: [0xa2,0x05,0xe3,0xff]
+	vabal.u32	q8, d19, d18
+
diff --git a/test/MC/ARM/neont2-bitcount-encoding.s b/test/MC/ARM/neont2-bitcount-encoding.s
new file mode 100644
index 0000000..cebd972
--- /dev/null
+++ b/test/MC/ARM/neont2-bitcount-encoding.s
@@ -0,0 +1,34 @@
+@ RUN: llvm-mc -mcpu=cortex-a8 -triple thumb-unknown-unkown -show-encoding < %s | FileCheck %s
+@ XFAIL: *
+
+.code 16
+
+@ CHECK: vcnt.8	d16, d16                @ encoding: [0x20,0x05,0xf0,0xff]
+	vcnt.8	d16, d16
+@ CHECK: vcnt.8	q8, q8                  @ encoding: [0x60,0x05,0xf0,0xff]
+	vcnt.8	q8, q8
+@ CHECK: vclz.i8	d16, d16                @ encoding: [0xa0,0x04,0xf0,0xff]
+	vclz.i8	d16, d16
+@ CHECK: vclz.i16	d16, d16        @ encoding: [0xa0,0x04,0xf4,0xff]
+	vclz.i16	d16, d16
+@ CHECK: vclz.i32	d16, d16        @ encoding: [0xa0,0x04,0xf8,0xff]
+	vclz.i32	d16, d16
+@ CHECK: vclz.i8	q8, q8                  @ encoding: [0xe0,0x04,0xf0,0xff]
+	vclz.i8	q8, q8
+@ CHECK: vclz.i16	q8, q8          @ encoding: [0xe0,0x04,0xf4,0xff]
+	vclz.i16	q8, q8
+@ CHECK: vclz.i32	q8, q8          @ encoding: [0xe0,0x04,0xf8,0xff]
+	vclz.i32	q8, q8
+@ CHECK: vcls.s8	d16, d16                @ encoding: [0x20,0x04,0xf0,0xff]
+	vcls.s8	d16, d16
+@ CHECK: vcls.s16	d16, d16        @ encoding: [0x20,0x04,0xf4,0xff]
+	vcls.s16	d16, d16
+@ CHECK: vcls.s32	d16, d16        @ encoding: [0x20,0x04,0xf8,0xff]
+	vcls.s32	d16, d16
+@ CHECK: vcls.s8	q8, q8                  @ encoding: [0x60,0x04,0xf0,0xff]
+	vcls.s8	q8, q8
+@ CHECK: vcls.s16	q8, q8          @ encoding: [0x60,0x04,0xf4,0xff]
+	vcls.s16	q8, q8
+@ CHECK: vcls.s32	q8, q8          @ encoding: [0x60,0x04,0xf8,0xff]
+	vcls.s32	q8, q8
+
diff --git a/test/MC/ARM/neont2-bitwise-encoding.s b/test/MC/ARM/neont2-bitwise-encoding.s
new file mode 100644
index 0000000..f5062f8
--- /dev/null
+++ b/test/MC/ARM/neont2-bitwise-encoding.s
@@ -0,0 +1,49 @@
+@ RUN: llvm-mc -mcpu=cortex-a8 -triple thumb-unknown-unkown -show-encoding < %s | FileCheck %s
+@ XFAIL: *
+
+.code 16
+
+@ CHECK: vand	d16, d17, d16           @ encoding: [0xb0,0x01,0x41,0xef]
+	vand	d16, d17, d16
+@ CHECK: vand	q8, q8, q9              @ encoding: [0xf2,0x01,0x40,0xef]
+	vand	q8, q8, q9
+
+@ CHECK: veor	d16, d17, d16           @ encoding: [0xb0,0x01,0x41,0xff]
+	veor	d16, d17, d16
+@ CHECK: veor	q8, q8, q9              @ encoding: [0xf2,0x01,0x40,0xff]
+	veor	q8, q8, q9
+
+@ CHECK: vorr	d16, d17, d16           @ encoding: [0xb0,0x01,0x61,0xef]
+	vorr	d16, d17, d16
+@ CHECK: vorr	q8, q8, q9              @ encoding: [0xf2,0x01,0x60,0xef]
+	vorr	q8, q8, q9
+@ CHECK: vorr.i32	d16, #0x1000000 @ encoding: [0x11,0x07,0xc0,0xef]
+  vorr.i32	d16, #0x1000000
+@ CHECK: vorr.i32	q8, #0x1000000  @ encoding: [0x51,0x07,0xc0,0xef]
+  vorr.i32	q8, #0x1000000
+@ CHECK: vorr.i32	q8, #0x0        @ encoding: [0x50,0x01,0xc0,0xef]
+  vorr.i32	q8, #0x0
+
+@ CHECK: vbic	d16, d17, d16           @ encoding: [0xb0,0x01,0x51,0xef]
+	vbic	d16, d17, d16
+@ CHECK: vbic	q8, q8, q9              @ encoding: [0xf2,0x01,0x50,0xef]
+	vbic	q8, q8, q9
+@ CHECK: vbic.i32	d16, #0xFF000000 @ encoding: [0x3f,0x07,0xc7,0xff]
+  vbic.i32	d16, #0xFF000000
+@ CHECK: vbic.i32	q8, #0xFF000000 @ encoding: [0x7f,0x07,0xc7,0xff]
+  vbic.i32	q8, #0xFF000000
+
+@ CHECK: vorn	d16, d17, d16           @ encoding: [0xb0,0x01,0x71,0xef]
+	vorn	d16, d17, d16
+@ CHECK: vorn	q8, q8, q9              @ encoding: [0xf2,0x01,0x70,0xef]
+	vorn	q8, q8, q9
+
+@ CHECK: vmvn	d16, d16                @ encoding: [0xa0,0x05,0xf0,0xff]
+	vmvn	d16, d16
+@ CHECK: vmvn	q8, q8                  @ encoding: [0xe0,0x05,0xf0,0xff]
+	vmvn	q8, q8
+
+@ CHECK: vbsl	d18, d17, d16           @ encoding: [0xb0,0x21,0x51,0xff]
+	vbsl	d18, d17, d16
+@ CHECK: vbsl	q8, q10, q9             @ encoding: [0xf2,0x01,0x54,0xff]
+	vbsl	q8, q10, q9
diff --git a/test/MC/ARM/neont2-convert-encoding.s b/test/MC/ARM/neont2-convert-encoding.s
new file mode 100644
index 0000000..9b00960
--- /dev/null
+++ b/test/MC/ARM/neont2-convert-encoding.s
@@ -0,0 +1,36 @@
+@ RUN: llvm-mc -mcpu=cortex-a8 -triple thumb-unknown-unkown -show-encoding < %s | FileCheck %s
+
+.code 16
+
+@ CHECK: vcvt.s32.f32	d16, d16        @ encoding: [0x20,0x07,0xfb,0xff]
+	vcvt.s32.f32	d16, d16
+@ CHECK: vcvt.u32.f32	d16, d16        @ encoding: [0xa0,0x07,0xfb,0xff]
+	vcvt.u32.f32	d16, d16
+@ CHECK: vcvt.f32.s32	d16, d16        @ encoding: [0x20,0x06,0xfb,0xff]
+	vcvt.f32.s32	d16, d16
+@ CHECK: vcvt.f32.u32	d16, d16        @ encoding: [0xa0,0x06,0xfb,0xff]
+	vcvt.f32.u32	d16, d16
+@ CHECK: vcvt.s32.f32	q8, q8          @ encoding: [0x60,0x07,0xfb,0xff]
+	vcvt.s32.f32	q8, q8
+@ CHECK: vcvt.u32.f32	q8, q8          @ encoding: [0xe0,0x07,0xfb,0xff]
+	vcvt.u32.f32	q8, q8
+@ CHECK: vcvt.f32.s32	q8, q8          @ encoding: [0x60,0x06,0xfb,0xff]
+	vcvt.f32.s32	q8, q8
+@ CHECK: vcvt.f32.u32	q8, q8          @ encoding: [0xe0,0x06,0xfb,0xff]
+	vcvt.f32.u32	q8, q8
+@ CHECK: vcvt.s32.f32	d16, d16, #1    @ encoding: [0x30,0x0f,0xff,0xef]
+	vcvt.s32.f32	d16, d16, #1
+@ CHECK: vcvt.u32.f32	d16, d16, #1    @ encoding: [0x30,0x0f,0xff,0xff]
+	vcvt.u32.f32	d16, d16, #1
+@ CHECK: vcvt.f32.s32	d16, d16, #1    @ encoding: [0x30,0x0e,0xff,0xef]
+	vcvt.f32.s32	d16, d16, #1
+@ CHECK: vcvt.f32.u32	d16, d16, #1    @ encoding: [0x30,0x0e,0xff,0xff]
+	vcvt.f32.u32	d16, d16, #1
+@ CHECK: vcvt.s32.f32	q8, q8, #1      @ encoding: [0x70,0x0f,0xff,0xef]
+	vcvt.s32.f32	q8, q8, #1
+@ CHECK: vcvt.u32.f32	q8, q8, #1      @ encoding: [0x70,0x0f,0xff,0xff]
+	vcvt.u32.f32	q8, q8, #1
+@ CHECK: vcvt.f32.s32	q8, q8, #1      @ encoding: [0x70,0x0e,0xff,0xef]
+	vcvt.f32.s32	q8, q8, #1
+@ CHECK: vcvt.f32.u32	q8, q8, #1      @ encoding: [0x70,0x0e,0xff,0xff]
+	vcvt.f32.u32	q8, q8, #1
diff --git a/test/MC/ARM/neont2-dup-encoding.s b/test/MC/ARM/neont2-dup-encoding.s
new file mode 100644
index 0000000..5330833
--- /dev/null
+++ b/test/MC/ARM/neont2-dup-encoding.s
@@ -0,0 +1,29 @@
+@ RUN: llvm-mc -mcpu=cortex-a8 -triple thumb-unknown-unkown -show-encoding < %s | FileCheck %s
+@ XFAIL: *
+
+.code 16
+
+@ CHECK: vdup.8	d16, r0                 @ encoding: [0x90,0x0b,0xc0,0xee]
+	vdup.8	d16, r0
+@ CHECK: vdup.16	d16, r0                 @ encoding: [0xb0,0x0b,0x80,0xee]
+	vdup.16	d16, r0
+@ CHECK: vdup.32	d16, r0                 @ encoding: [0x90,0x0b,0x80,0xee]
+	vdup.32	d16, r0
+@ CHECK: vdup.8	q8, r0                  @ encoding: [0x90,0x0b,0xe0,0xee]
+	vdup.8	q8, r0
+@ CHECK: vdup.16	q8, r0                  @ encoding: [0xb0,0x0b,0xa0,0xee]
+	vdup.16	q8, r0
+@ CHECK: vdup.32	q8, r0                  @ encoding: [0x90,0x0b,0xa0,0xee]
+	vdup.32	q8, r0
+@ CHECK: vdup.8	d16, d16[1]             @ encoding: [0x20,0x0c,0xf3,0xff]
+	vdup.8	d16, d16[1]
+@ CHECK: vdup.16	d16, d16[1]             @ encoding: [0x20,0x0c,0xf6,0xff]
+	vdup.16	d16, d16[1]
+@ CHECK: vdup.32	d16, d16[1]             @ encoding: [0x20,0x0c,0xfc,0xff]
+	vdup.32	d16, d16[1]
+@ CHECK: vdup.8	q8, d16[1]              @ encoding: [0x60,0x0c,0xf3,0xff]
+	vdup.8	q8, d16[1]
+@ CHECK: vdup.16	q8, d16[1]              @ encoding: [0x60,0x0c,0xf6,0xff]
+	vdup.16	q8, d16[1]
+@ CHECK: vdup.32	q8, d16[1]              @ encoding: [0x60,0x0c,0xfc,0xff]
+	vdup.32	q8, d16[1]
diff --git a/test/MC/ARM/neont2-mov-encoding.s b/test/MC/ARM/neont2-mov-encoding.s
new file mode 100644
index 0000000..4f6b687
--- /dev/null
+++ b/test/MC/ARM/neont2-mov-encoding.s
@@ -0,0 +1,119 @@
+@ RUN: llvm-mc -mcpu=cortex-a8 -triple thumb-unknown-unkown -show-encoding < %s | FileCheck %s
+@ XFAIL: *
+
+.code 16
+
+@ CHECK: vmov.i8	d16, #0x8               @ encoding: [0x18,0x0e,0xc0,0xef]
+	vmov.i8	d16, #0x8
+@ CHECK: vmov.i16	d16, #0x10      @ encoding: [0x10,0x08,0xc1,0xef]
+	vmov.i16	d16, #0x10
+@ CHECK: vmov.i16	d16, #0x1000    @ encoding: [0x10,0x0a,0xc1,0xef]
+	vmov.i16	d16, #0x1000
+@ CHECK: vmov.i32	d16, #0x20      @ encoding: [0x10,0x00,0xc2,0xef]
+	vmov.i32	d16, #0x20
+@ CHECK: vmov.i32	d16, #0x2000    @ encoding: [0x10,0x02,0xc2,0xef]
+	vmov.i32	d16, #0x2000
+@ CHECK: vmov.i32	d16, #0x200000  @ encoding: [0x10,0x04,0xc2,0xef]
+	vmov.i32	d16, #0x200000
+@ CHECK: vmov.i32	d16, #0x20000000 @ encoding: [0x10,0x06,0xc2,0xef]
+	vmov.i32	d16, #0x20000000
+@ CHECK: vmov.i32	d16, #0x20FF    @ encoding: [0x10,0x0c,0xc2,0xef]
+	vmov.i32	d16, #0x20FF
+@ CHECK: vmov.i32	d16, #0x20FFFF  @ encoding: [0x10,0x0d,0xc2,0xef]
+	vmov.i32	d16, #0x20FFFF
+@ CHECK: vmov.i64	d16, #0xFF0000FF0000FFFF @ encoding: [0x33,0x0e,0xc1,0xff]
+	vmov.i64	d16, #0xFF0000FF0000FFFF
+@ CHECK: vmov.i8	q8, #0x8                @ encoding: [0x58,0x0e,0xc0,0xef]
+	vmov.i8	q8, #0x8
+@ CHECK: vmov.i16	q8, #0x10       @ encoding: [0x50,0x08,0xc1,0xef]
+	vmov.i16	q8, #0x10
+@ CHECK: vmov.i16	q8, #0x1000     @ encoding: [0x50,0x0a,0xc1,0xef]
+	vmov.i16	q8, #0x1000
+@ CHECK: vmov.i32	q8, #0x20       @ encoding: [0x50,0x00,0xc2,0xef]
+	vmov.i32	q8, #0x20
+@ CHECK: vmov.i32	q8, #0x2000     @ encoding: [0x50,0x02,0xc2,0xef]
+	vmov.i32	q8, #0x2000
+@ CHECK: vmov.i32	q8, #0x200000   @ encoding: [0x50,0x04,0xc2,0xef]
+	vmov.i32	q8, #0x200000
+@ CHECK: vmov.i32	q8, #0x20000000 @ encoding: [0x50,0x06,0xc2,0xef]
+	vmov.i32	q8, #0x20000000
+@ CHECK: vmov.i32	q8, #0x20FF     @ encoding: [0x50,0x0c,0xc2,0xef]
+	vmov.i32	q8, #0x20FF
+@ CHECK: vmov.i32	q8, #0x20FFFF   @ encoding: [0x50,0x0d,0xc2,0xef]
+	vmov.i32	q8, #0x20FFFF
+@ CHECK: vmov.i64	q8, #0xFF0000FF0000FFFF @ encoding: [0x73,0x0e,0xc1,0xff]
+	vmov.i64	q8, #0xFF0000FF0000FFFF
+@ CHECK: vmvn.i16	d16, #0x10      @ encoding: [0x30,0x08,0xc1,0xef]
+	vmvn.i16	d16, #0x10
+@ CHECK: vmvn.i16	d16, #0x1000    @ encoding: [0x30,0x0a,0xc1,0xef]
+	vmvn.i16	d16, #0x1000
+@ CHECK: vmvn.i32	d16, #0x20      @ encoding: [0x30,0x00,0xc2,0xef]
+	vmvn.i32	d16, #0x20
+@ CHECK: vmvn.i32	d16, #0x2000    @ encoding: [0x30,0x02,0xc2,0xef]
+	vmvn.i32	d16, #0x2000
+@ CHECK: vmvn.i32	d16, #0x200000  @ encoding: [0x30,0x04,0xc2,0xef]
+	vmvn.i32	d16, #0x200000
+@ CHECK: vmvn.i32	d16, #0x20000000 @ encoding: [0x30,0x06,0xc2,0xef]
+	vmvn.i32	d16, #0x20000000
+@ CHECK: vmvn.i32	d16, #0x20FF    @ encoding: [0x30,0x0c,0xc2,0xef]
+	vmvn.i32	d16, #0x20FF
+@ CHECK: vmvn.i32	d16, #0x20FFFF  @ encoding: [0x30,0x0d,0xc2,0xef]
+	vmvn.i32	d16, #0x20FFFF
+@ CHECK: vmovl.s8	q8, d16         @ encoding: [0x30,0x0a,0xc8,0xef]
+	vmovl.s8	q8, d16
+@ CHECK: vmovl.s16	q8, d16         @ encoding: [0x30,0x0a,0xd0,0xef]
+	vmovl.s16	q8, d16
+@ CHECK: vmovl.s32	q8, d16         @ encoding: [0x30,0x0a,0xe0,0xef]
+	vmovl.s32	q8, d16
+@ CHECK: vmovl.u8	q8, d16         @ encoding: [0x30,0x0a,0xc8,0xff]
+	vmovl.u8	q8, d16
+@ CHECK: vmovl.u16	q8, d16         @ encoding: [0x30,0x0a,0xd0,0xff]
+	vmovl.u16	q8, d16
+@ CHECK: vmovl.u32	q8, d16         @ encoding: [0x30,0x0a,0xe0,0xff]
+	vmovl.u32	q8, d16
+@ CHECK: vmovn.i16	d16, q8         @ encoding: [0x20,0x02,0xf2,0xff]
+	vmovn.i16	d16, q8
+@ CHECK: vmovn.i32	d16, q8         @ encoding: [0x20,0x02,0xf6,0xff]
+	vmovn.i32	d16, q8
+@ CHECK: vmovn.i64	d16, q8         @ encoding: [0x20,0x02,0xfa,0xff]
+	vmovn.i64	d16, q8
+@ CHECK: vqmovn.s16	d16, q8         @ encoding: [0xa0,0x02,0xf2,0xff]
+	vqmovn.s16	d16, q8
+@ CHECK: vqmovn.s32	d16, q8         @ encoding: [0xa0,0x02,0xf6,0xff]
+	vqmovn.s32	d16, q8
+@ CHECK: vqmovn.s64	d16, q8         @ encoding: [0xa0,0x02,0xfa,0xff]
+	vqmovn.s64	d16, q8
+@ CHECK: vqmovn.u16	d16, q8         @ encoding: [0xe0,0x02,0xf2,0xff]
+	vqmovn.u16	d16, q8
+@ CHECK: vqmovn.u32	d16, q8         @ encoding: [0xe0,0x02,0xf6,0xff]
+	vqmovn.u32	d16, q8
+@ CHECK: vqmovn.u64	d16, q8         @ encoding: [0xe0,0x02,0xfa,0xff]
+	vqmovn.u64	d16, q8
+@ CHECK: vqmovun.s16	d16, q8         @ encoding: [0x60,0x02,0xf2,0xff]
+	vqmovun.s16	d16, q8
+@ CHECK: vqmovun.s32	d16, q8         @ encoding: [0x60,0x02,0xf6,0xff]
+	vqmovun.s32	d16, q8
+@ CHECK: vqmovun.s64	d16, q8         @ encoding: [0x60,0x02,0xfa,0xff]
+	vqmovun.s64	d16, q8
+@ CHECK: vmov.s8	r0, d16[1]              @ encoding: [0xb0,0x0b,0x50,0xee]
+	vmov.s8	r0, d16[1]
+@ CHECK: vmov.s16	r0, d16[1]      @ encoding: [0xf0,0x0b,0x10,0xee]
+	vmov.s16	r0, d16[1]
+@ CHECK: vmov.u8	r0, d16[1]              @ encoding: [0xb0,0x0b,0xd0,0xee]
+	vmov.u8	r0, d16[1]
+@ CHECK: vmov.u16	r0, d16[1]      @ encoding: [0xf0,0x0b,0x90,0xee]
+	vmov.u16	r0, d16[1]
+@ CHECK: vmov.32	r0, d16[1]              @ encoding: [0x90,0x0b,0x30,0xee]
+	vmov.32	r0, d16[1]
+@ CHECK: vmov.8	d16[1], r1              @ encoding: [0xb0,0x1b,0x40,0xee]
+	vmov.8	d16[1], r1
+@ CHECK: vmov.16	d16[1], r1              @ encoding: [0xf0,0x1b,0x00,0xee]
+	vmov.16	d16[1], r1
+@ CHECK: vmov.32	d16[1], r1              @ encoding: [0x90,0x1b,0x20,0xee]
+	vmov.32	d16[1], r1
+@ CHECK: vmov.8	d18[1], r1              @ encoding: [0xb0,0x1b,0x42,0xee]
+	vmov.8	d18[1], r1
+@ CHECK: vmov.16	d18[1], r1              @ encoding: [0xf0,0x1b,0x02,0xee]
+	vmov.16	d18[1], r1
+@ CHECK: vmov.32	d18[1], r1              @ encoding: [0x90,0x1b,0x22,0xee]
+	vmov.32	d18[1], r1
diff --git a/test/MC/ARM/neont2-mul-accum-encoding.s b/test/MC/ARM/neont2-mul-accum-encoding.s
new file mode 100644
index 0000000..66d4e10
--- /dev/null
+++ b/test/MC/ARM/neont2-mul-accum-encoding.s
@@ -0,0 +1,69 @@
+@ RUN: llvm-mc -mcpu=cortex-a8 -triple thumb-unknown-unkown -show-encoding < %s | FileCheck %s
+@ XFAIL: *
+
+.code 16
+
+@ CHECK: vmla.i8	d16, d18, d17           @ encoding: [0xa1,0x09,0x42,0xef]
+	vmla.i8	d16, d18, d17
+@ CHECK: vmla.i16	d16, d18, d17   @ encoding: [0xa1,0x09,0x52,0xef]
+	vmla.i16	d16, d18, d17
+@ CHECK: vmla.i16	d16, d18, d17   @ encoding: [0xa1,0x09,0x52,0xef]
+	vmla.i32	d16, d18, d17
+@ CHECK: vmla.f32	d16, d18, d17   @ encoding: [0xb1,0x0d,0x42,0xef]
+	vmla.f32	d16, d18, d17
+@ CHECK: vmla.i8	q9, q8, q10             @ encoding: [0xe4,0x29,0x40,0xef]
+	vmla.i8	q9, q8, q10
+@ CHECK: vmla.i16	q9, q8, q10     @ encoding: [0xe4,0x29,0x50,0xef]
+	vmla.i16	q9, q8, q10
+@ CHECK: vmla.i32	q9, q8, q10     @ encoding: [0xe4,0x29,0x60,0xef]
+	vmla.i32	q9, q8, q10
+@ CHECK: vmla.f32	q9, q8, q10     @ encoding: [0xf4,0x2d,0x40,0xef]
+	vmla.f32	q9, q8, q10
+@ CHECK: vmlal.s8	q8, d19, d18    @ encoding: [0xa2,0x08,0xc3,0xef]
+	vmlal.s8	q8, d19, d18
+@ CHECK: vmlal.s16	q8, d19, d18    @ encoding: [0xa2,0x08,0xd3,0xef]
+	vmlal.s16	q8, d19, d18
+@ CHECK: vmlal.s32	q8, d19, d18    @ encoding: [0xa2,0x08,0xe3,0xef]
+	vmlal.s32	q8, d19, d18
+@ CHECK: vmlal.u8	q8, d19, d18    @ encoding: [0xa2,0x08,0xc3,0xff]
+	vmlal.u8	q8, d19, d18
+@ CHECK: vmlal.u16	q8, d19, d18    @ encoding: [0xa2,0x08,0xd3,0xff]
+	vmlal.u16	q8, d19, d18
+@ CHECK: vmlal.u32	q8, d19, d18    @ encoding: [0xa2,0x08,0xe3,0xff]
+	vmlal.u32	q8, d19, d18
+@ CHECK: vqdmlal.s16	q8, d19, d18    @ encoding: [0xa2,0x09,0xd3,0xef]
+	vqdmlal.s16	q8, d19, d18
+@ CHECK: vqdmlal.s32	q8, d19, d18    @ encoding: [0xa2,0x09,0xe3,0xef]
+	vqdmlal.s32	q8, d19, d18
+@ CHECK: vmls.i8	d16, d18, d17           @ encoding: [0xa1,0x09,0x42,0xff]
+	vmls.i8	d16, d18, d17
+@ CHECK: vmls.i16	d16, d18, d17   @ encoding: [0xa1,0x09,0x52,0xff]
+	vmls.i16	d16, d18, d17
+@ CHECK: vmls.i32	d16, d18, d17   @ encoding: [0xa1,0x09,0x62,0xff]
+	vmls.i32	d16, d18, d17
+@ CHECK: vmls.f32	d16, d18, d17   @ encoding: [0xb1,0x0d,0x62,0xef]
+	vmls.f32	d16, d18, d17
+@ CHECK: vmls.i8	q9, q8, q10             @ encoding: [0xe4,0x29,0x40,0xff]
+	vmls.i8	q9, q8, q10
+@ CHECK: vmls.i16	q9, q8, q10     @ encoding: [0xe4,0x29,0x50,0xff]
+	vmls.i16	q9, q8, q10
+@ CHECK: vmls.i32	q9, q8, q10     @ encoding: [0xe4,0x29,0x60,0xff]
+	vmls.i32	q9, q8, q10
+@ CHECK: vmls.f32	q9, q8, q10     @ encoding: [0xf4,0x2d,0x60,0xef]
+	vmls.f32	q9, q8, q10
+@ CHECK: vmlsl.s8	q8, d19, d18    @ encoding: [0xa2,0x0a,0xc3,0xef]
+	vmlsl.s8	q8, d19, d18
+@ CHECK: vmlsl.s16	q8, d19, d18    @ encoding: [0xa2,0x0a,0xd3,0xef]
+	vmlsl.s16	q8, d19, d18
+@ CHECK: vmlsl.s32	q8, d19, d18    @ encoding: [0xa2,0x0a,0xe3,0xef]
+	vmlsl.s32	q8, d19, d18
+@ CHECK: vmlsl.u8	q8, d19, d18    @ encoding: [0xa2,0x0a,0xc3,0xff]
+	vmlsl.u8	q8, d19, d18
+@ CHECK: vmlsl.u16	q8, d19, d18    @ encoding: [0xa2,0x0a,0xd3,0xff]
+	vmlsl.u16	q8, d19, d18
+@ CHECK: vmlsl.u32	q8, d19, d18    @ encoding: [0xa2,0x0a,0xe3,0xff]
+	vmlsl.u32	q8, d19, d18
+@ CHECK: vqdmlsl.s16	q8, d19, d18    @ encoding: [0xa2,0x0b,0xd3,0xef]
+	vqdmlsl.s16	q8, d19, d18
+@ CHECK: vqdmlsl.s32	q8, d19, d18    @ encoding: [0xa2,0x0b,0xe3,0xef]
+	vqdmlsl.s32	q8, d19, d18
diff --git a/test/MC/ARM/neont2-pairwise-encoding.s b/test/MC/ARM/neont2-pairwise-encoding.s
new file mode 100644
index 0000000..058f1d8
--- /dev/null
+++ b/test/MC/ARM/neont2-pairwise-encoding.s
@@ -0,0 +1,89 @@
+@ RUN: llvm-mc -mcpu=cortex-a8 -triple thumb-unknown-unkown -show-encoding < %s | FileCheck %s
+@ XFAIL: *
+
+.code 16
+
+@ CHECK: vpadd.i8	d16, d17, d16   @ encoding: [0xb0,0x0b,0x41,0xef]
+	vpadd.i8	d16, d17, d16
+@ CHECK: vpadd.i16	d16, d17, d16   @ encoding: [0xb0,0x0b,0x51,0xef]
+	vpadd.i16	d16, d17, d16
+@ CHECK: vpadd.i32	d16, d17, d16   @ encoding: [0xb0,0x0b,0x61,0xef]
+	vpadd.i32	d16, d17, d16
+@ CHECK: vpadd.f32	d16, d16, d17   @ encoding: [0xa1,0x0d,0x40,0xff]
+	vpadd.f32	d16, d16, d17
+@ CHECK: vpaddl.s8	d16, d16        @ encoding: [0x20,0x02,0xf0,0xff]
+	vpaddl.s8	d16, d16
+@ CHECK: vpaddl.s16	d16, d16        @ encoding: [0x20,0x02,0xf4,0xff]
+	vpaddl.s16	d16, d16
+@ CHECK: vpaddl.s32	d16, d16        @ encoding: [0x20,0x02,0xf8,0xff]
+	vpaddl.s32	d16, d16
+@ CHECK: vpaddl.u8	d16, d16        @ encoding: [0xa0,0x02,0xf0,0xff]
+	vpaddl.u8	d16, d16
+@ CHECK: vpaddl.u16	d16, d16        @ encoding: [0xa0,0x02,0xf4,0xff]
+	vpaddl.u16	d16, d16
+@ CHECK: vpaddl.u32	d16, d16        @ encoding: [0xa0,0x02,0xf8,0xff]
+	vpaddl.u32	d16, d16
+@ CHECK: vpaddl.s8	q8, q8          @ encoding: [0x60,0x02,0xf0,0xff]
+	vpaddl.s8	q8, q8
+@ CHECK: vpaddl.s16	q8, q8          @ encoding: [0x60,0x02,0xf4,0xff]
+	vpaddl.s16	q8, q8
+@ CHECK: vpaddl.s32	q8, q8          @ encoding: [0x60,0x02,0xf8,0xff]
+	vpaddl.s32	q8, q8
+@ CHECK: vpaddl.u8	q8, q8          @ encoding: [0xe0,0x02,0xf0,0xff]
+	vpaddl.u8	q8, q8
+@ CHECK: vpaddl.u16	q8, q8          @ encoding: [0xe0,0x02,0xf4,0xff]
+	vpaddl.u16	q8, q8
+@ CHECK: vpaddl.u32	q8, q8          @ encoding: [0xe0,0x02,0xf8,0xff]
+	vpaddl.u32	q8, q8
+@ CHECK: vpadal.s8	d16, d17        @ encoding: [0x21,0x06,0xf0,0xff]
+	vpadal.s8	d16, d17
+@ CHECK: vpadal.s16	d16, d17        @ encoding: [0x21,0x06,0xf4,0xff]
+	vpadal.s16	d16, d17
+@ CHECK: vpadal.s32	d16, d17        @ encoding: [0x21,0x06,0xf8,0xff]
+	vpadal.s32	d16, d17
+@ CHECK: vpadal.u8	d16, d17        @ encoding: [0xa1,0x06,0xf0,0xff]
+	vpadal.u8	d16, d17
+@ CHECK: vpadal.u16	d16, d17        @ encoding: [0xa1,0x06,0xf4,0xff]
+	vpadal.u16	d16, d17
+@ CHECK: vpadal.u32	d16, d17        @ encoding: [0xa1,0x06,0xf8,0xff]
+	vpadal.u32	d16, d17
+@ CHECK: vpadal.s8	q9, q8          @ encoding: [0x60,0x26,0xf0,0xff]
+	vpadal.s8	q9, q8
+@ CHECK: vpadal.s16	q9, q8          @ encoding: [0x60,0x26,0xf4,0xff]
+	vpadal.s16	q9, q8
+@ CHECK: vpadal.s32	q9, q8          @ encoding: [0x60,0x26,0xf8,0xff]
+	vpadal.s32	q9, q8
+@ CHECK: vpadal.u8	q9, q8          @ encoding: [0xe0,0x26,0xf0,0xff]
+	vpadal.u8	q9, q8
+@ CHECK: vpadal.u16	q9, q8          @ encoding: [0xe0,0x26,0xf4,0xff]
+	vpadal.u16	q9, q8
+@ CHECK: vpadal.u32	q9, q8          @ encoding: [0xe0,0x26,0xf8,0xff]
+	vpadal.u32	q9, q8
+@ CHECK: vpmin.s8	d16, d16, d17   @ encoding: [0xb1,0x0a,0x40,0xef]
+	vpmin.s8	d16, d16, d17
+@ CHECK: vpmin.s16	d16, d16, d17   @ encoding: [0xb1,0x0a,0x50,0xef]
+	vpmin.s16	d16, d16, d17
+@ CHECK: vpmin.s32	d16, d16, d17   @ encoding: [0xb1,0x0a,0x60,0xef]
+	vpmin.s32	d16, d16, d17
+@ CHECK: vpmin.u8	d16, d16, d17   @ encoding: [0xb1,0x0a,0x40,0xff]
+	vpmin.u8	d16, d16, d17
+@ CHECK: vpmin.u16	d16, d16, d17   @ encoding: [0xb1,0x0a,0x50,0xff]
+	vpmin.u16	d16, d16, d17
+@ CHECK: vpmin.u32	d16, d16, d17   @ encoding: [0xb1,0x0a,0x60,0xff]
+	vpmin.u32	d16, d16, d17
+@ CHECK: vpmin.f32	d16, d16, d17   @ encoding: [0xa1,0x0f,0x60,0xff]
+	vpmin.f32	d16, d16, d17
+@ CHECK: vpmax.s8	d16, d16, d17   @ encoding: [0xa1,0x0a,0x40,0xef]
+	vpmax.s8	d16, d16, d17
+@ CHECK: vpmax.s16	d16, d16, d17   @ encoding: [0xa1,0x0a,0x50,0xef]
+	vpmax.s16	d16, d16, d17
+@ CHECK: vpmax.s32	d16, d16, d17   @ encoding: [0xa1,0x0a,0x60,0xef]
+	vpmax.s32	d16, d16, d17
+@ CHECK: vpmax.u8	d16, d16, d17   @ encoding: [0xa1,0x0a,0x40,0xff]
+	vpmax.u8	d16, d16, d17
+@ CHECK: vpmax.u16	d16, d16, d17   @ encoding: [0xa1,0x0a,0x50,0xff]
+	vpmax.u16	d16, d16, d17
+@ CHECK: vpmax.u32	d16, d16, d17   @ encoding: [0xa1,0x0a,0x60,0xff]
+	vpmax.u32	d16, d16, d17
+@ CHECK: vpmax.f32	d16, d16, d17   @ encoding: [0xa1,0x0f,0x40,0xff]
+	vpmax.f32	d16, d16, d17
diff --git a/test/MC/ARM/neont2-table-encoding.s b/test/MC/ARM/neont2-table-encoding.s
new file mode 100644
index 0000000..3589f2f
--- /dev/null
+++ b/test/MC/ARM/neont2-table-encoding.s
@@ -0,0 +1,21 @@
+@ RUN: llvm-mc -mcpu=cortex-a8 -triple thumb-unknown-unkown -show-encoding < %s | FileCheck %s
+@ XFAIL: *
+
+.code 16
+
+@ CHECK: vtbl.8	d16, {d17}, d16         @ encoding: [0xa0,0x08,0xf1,0xff]
+	vtbl.8	d16, {d17}, d16
+@ CHECK: vtbl.8	d16, {d16, d17}, d18    @ encoding: [0xa2,0x09,0xf0,0xff]
+	vtbl.8	d16, {d16, d17}, d18
+@ CHECK: vtbl.8	d16, {d16, d17, d18}, d20 @ encoding: [0xa4,0x0a,0xf0,0xff]
+	vtbl.8	d16, {d16, d17, d18}, d20
+@ CHECK: vtbl.8	d16, {d16, d17, d18, d19}, d20 @ encoding: [0xa4,0x0b,0xf0,0xff]
+	vtbl.8	d16, {d16, d17, d18, d19}, d20
+@ CHECK: vtbx.8	d18, {d16}, d17         @ encoding: [0xe1,0x28,0xf0,0xff]
+	vtbx.8	d18, {d16}, d17
+@ CHECK: vtbx.8	d19, {d16, d17}, d18    @ encoding: [0xe2,0x39,0xf0,0xff]
+	vtbx.8	d19, {d16, d17}, d18
+@ CHECK: vtbx.8	d20, {d16, d17, d18}, d21 @ encoding: [0xe5,0x4a,0xf0,0xff]
+	vtbx.8	d20, {d16, d17, d18}, d21
+@ CHECK: vtbx.8	d20, {d16, d17, d18, d19}, d21 @ encoding: [0xe5,0x4b,0xf0,0xff]
+	vtbx.8	d20, {d16, d17, d18, d19}, d21