-G# RUN: llvm-mc -triple aarch64-none-linux-gnu -mattr=+neon -disassemble < %s | FileCheck %s
+# RUN: llvm-mc -triple aarch64-none-linux-gnu -mattr=+neon -disassemble < %s | FileCheck %s
+# RUN: llvm-mc -triple arm64-none-linux-gnu -mattr=+neon -disassemble < %s | FileCheck %s
#------------------------------------------------------------------------------
# Vector Integer Add/Sub
# Vector Bitwise OR - immedidate
#------------------------------------------------------------------------------
# CHECK: movi v31.4s, #0xff, lsl #24
-# CHECK: mvni v0.2s, #0x0
+# CHECK: mvni v0.2s, #{{0x0|0}}
# CHECK: bic v15.4h, #0xf, lsl #8
# CHECK: orr v16.8h, #0x1f
0xff 0x67 0x07 0x4f
# Vector Move - register
#------------------------------------------------------------------------------
-# FIXME: these should print as "mov", but TableGen can't handle it.
-
-# CHECK: orr v1.16b, v15.16b, v15.16b
-# CHECK: orr v25.8b, v4.8b, v4.8b
+# CHECK: mov v1.16b, v15.16b
+# CHECK: mov v25.8b, v4.8b
0xe1 0x1d 0xaf 0x4e
0x99 0x1c 0xa4 0x0e
#----------------------------------------------------------------------
# Vector Compare Mask Equal to Zero (Integer)
#----------------------------------------------------------------------
-# CHECK: cmeq v31.16b, v15.16b, #0x0
+# CHECK: cmeq v31.16b, v15.16b, #{{0x0|0}}
0xff 0x99 0x20 0x4e
#----------------------------------------------------------------------
# Vector Compare Mask Greater Than or Equal to Zero (Signed Integer)
#----------------------------------------------------------------------
-# CHECK: cmge v3.8b, v15.8b, #0x0
+# CHECK: cmge v3.8b, v15.8b, #{{0x0|0}}
0xe3 0x89 0x20 0x2e
#----------------------------------------------------------------------
# Vector Compare Mask Greater Than Zero (Signed Integer)
#----------------------------------------------------------------------
-# CHECK: cmgt v22.2s, v9.2s, #0x0
+# CHECK: cmgt v22.2s, v9.2s, #{{0x0|0}}
0x36 0x89 0xa0 0x0e
#----------------------------------------------------------------------
# Vector Compare Mask Less Than or Equal To Zero (Signed Integer)
#----------------------------------------------------------------------
-# CHECK: cmle v5.2d, v14.2d, #0x0
+# CHECK: cmle v5.2d, v14.2d, #{{0x0|0}}
0xc5 0x99 0xe0 0x6e
#----------------------------------------------------------------------
# Vector Compare Mask Less Than Zero (Signed Integer)
#----------------------------------------------------------------------
-# CHECK: cmlt v13.8h, v11.8h, #0x0
+# CHECK: cmlt v13.8h, v11.8h, #{{0x0|0}}
0x6d 0xa9 0x60 0x4e
#----------------------------------------------------------------------
0xf5 0xdd 0x23 0x4e
0xab 0xdc 0x77 0x4e
+#----------------------------------------------------------------------
+# Vector Shift Left long
+#----------------------------------------------------------------------
+# CHECK: shll2 v2.8h, v4.16b, #8
+# CHECK: shll2 v6.4s, v8.8h, #16
+# CHECK: shll2 v6.2d, v8.4s, #32
+# CHECK: shll v2.8h, v4.8b, #8
+# CHECK: shll v6.4s, v8.4h, #16
+# CHECK: shll v6.2d, v8.2s, #32
+
+0x82,0x38,0x21,0x6e
+0x06,0x39,0x61,0x6e
+0x06,0x39,0xa1,0x6e
+0x82,0x38,0x21,0x2e
+0x06,0x39,0x61,0x2e
+0x06,0x39,0xa1,0x2e
+
#----------------------------------------------------------------------
# Vector Shift Left by Immediate
#----------------------------------------------------------------------
#----------------------------------------------------------------------
# Scalar Compare Bitwise Equal To Zero
#----------------------------------------------------------------------
-# CHECK: cmeq d20, d21, #0x0
+# CHECK: cmeq d20, d21, #{{0x0|0}}
0xb4,0x9a,0xe0,0x5e
#----------------------------------------------------------------------
#----------------------------------------------------------------------
# Scalar Compare Signed Greather Than Or Equal To Zero
#----------------------------------------------------------------------
-# CHECK: cmge d20, d21, #0x0
+# CHECK: cmge d20, d21, #{{0x0|0}}
0xb4,0x8a,0xe0,0x7e
#----------------------------------------------------------------------
#----------------------------------------------------------------------
# Scalar Compare Signed Greater Than Zero
#----------------------------------------------------------------------
-# CHECK: cmgt d20, d21, #0x0
+# CHECK: cmgt d20, d21, #{{0x0|0}}
0xb4,0x8a,0xe0,0x5e
#----------------------------------------------------------------------
# Scalar Compare Signed Less Than Or Equal To Zero
#----------------------------------------------------------------------
-# CHECK: cmle d20, d21, #0x0
+# CHECK: cmle d20, d21, #{{0x0|0}}
0xb4,0x9a,0xe0,0x7e
#----------------------------------------------------------------------
# Scalar Compare Less Than Zero
#----------------------------------------------------------------------
-# CHECK: cmlt d20, d21, #0x0
+# CHECK: cmlt d20, d21, #{{0x0|0}}
0xb4,0xaa,0xe0,0x5e
#----------------------------------------------------------------------
#----------------------------------------------------------------------
# Vector load/store multiple N-element structure
#----------------------------------------------------------------------
-# CHECK: ld1 {v0.16b}, [x0]
-# CHECK: ld1 {v15.8h, v16.8h}, [x15]
-# CHECK: ld1 {v31.4s, v0.4s, v1.4s}, [sp]
-# CHECK: ld1 {v0.2d, v1.2d, v2.2d, v3.2d}, [x0]
+# CHECK: ld1 { v0.16b }, [x0]
+# CHECK: ld1 { v15.8h, v16.8h }, [x15]
+# CHECK: ld1 { v31.4s, v0.4s, v1.4s }, [sp]
+# CHECK: ld1 { v0.2d, v1.2d, v2.2d, v3.2d }, [x0]
0x00,0x70,0x40,0x4c
0xef,0xa5,0x40,0x4c
0xff,0x6b,0x40,0x4c
0x00,0x2c,0x40,0x4c
-# CHECK: ld2 {v0.8b, v1.8b}, [x0]
-# CHECK: ld3 {v15.4h, v16.4h, v17.4h}, [x15]
-# CHECK: ld4 {v31.2s, v0.2s, v1.2s, v2.2s}, [sp]
+# CHECK: ld2 { v0.8b, v1.8b }, [x0]
+# CHECK: ld3 { v15.4h, v16.4h, v17.4h }, [x15]
+# CHECK: ld4 { v31.2s, v0.2s, v1.2s, v2.2s }, [sp]
0x00,0x80,0x40,0x0c
0xef,0x45,0x40,0x0c
0xff,0x0b,0x40,0x0c
-# CHECK: st1 {v0.16b}, [x0]
-# CHECK: st1 {v15.8h, v16.8h}, [x15]
-# CHECK: st1 {v31.4s, v0.4s, v1.4s}, [sp]
-# CHECK: st1 {v0.2d, v1.2d, v2.2d, v3.2d}, [x0]
+# CHECK: st1 { v0.16b }, [x0]
+# CHECK: st1 { v15.8h, v16.8h }, [x15]
+# CHECK: st1 { v31.4s, v0.4s, v1.4s }, [sp]
+# CHECK: st1 { v0.2d, v1.2d, v2.2d, v3.2d }, [x0]
0x00,0x70,0x00,0x4c
0xef,0xa5,0x00,0x4c
0xff,0x6b,0x00,0x4c
0x00,0x2c,0x00,0x4c
-# CHECK: st2 {v0.8b, v1.8b}, [x0]
-# CHECK: st3 {v15.4h, v16.4h, v17.4h}, [x15]
-# CHECK: st4 {v31.2s, v0.2s, v1.2s, v2.2s}, [sp]
+# CHECK: st2 { v0.8b, v1.8b }, [x0]
+# CHECK: st3 { v15.4h, v16.4h, v17.4h }, [x15]
+# CHECK: st4 { v31.2s, v0.2s, v1.2s, v2.2s }, [sp]
0x00,0x80,0x00,0x0c
0xef,0x45,0x00,0x0c
0xff,0x0b,0x00,0x0c
#----------------------------------------------------------------------
# Vector load/store multiple N-element structure (post-index)
#----------------------------------------------------------------------
-# CHECK: ld1 {v15.8h}, [x15], x2
-# CHECK: ld1 {v31.4s, v0.4s}, [sp], #32
-# CHECK: ld1 {v0.2d, v1.2d, v2.2d}, [x0], #48
-# CHECK: ld1 {v0.8b, v1.8b, v2.8b, v3.8b}, [x0], x3
+# CHECK: ld1 { v15.8h }, [x15], x2
+# CHECK: ld1 { v31.4s, v0.4s }, [sp], #32
+# CHECK: ld1 { v0.2d, v1.2d, v2.2d }, [x0], #48
+# CHECK: ld1 { v0.8b, v1.8b, v2.8b, v3.8b }, [x0], x3
0xef,0x75,0xc2,0x4c
0xff,0xab,0xdf,0x4c
0x00,0x6c,0xdf,0x4c
0x00,0x20,0xc3,0x0c
-# CHECK: ld2 {v0.16b, v1.16b}, [x0], x1
-# CHECK: ld3 {v15.8h, v16.8h, v17.8h}, [x15], x2
-# CHECK: ld4 {v31.4s, v0.4s, v1.4s, v2.4s}, [sp], #64
+# CHECK: ld2 { v0.16b, v1.16b }, [x0], x1
+# CHECK: ld3 { v15.8h, v16.8h, v17.8h }, [x15], x2
+# CHECK: ld4 { v31.4s, v0.4s, v1.4s, v2.4s }, [sp], #64
0x00,0x80,0xc1,0x4c
0xef,0x45,0xc2,0x4c
0xff,0x0b,0xdf,0x4c
-# CHECK: st1 {v15.8h}, [x15], x2
-# CHECK: st1 {v31.4s, v0.4s}, [sp], #32
-# CHECK: st1 {v0.2d, v1.2d, v2.2d}, [x0], #48
-# CHECK: st1 {v0.8b, v1.8b, v2.8b, v3.8b}, [x0], x3
+# CHECK: st1 { v15.8h }, [x15], x2
+# CHECK: st1 { v31.4s, v0.4s }, [sp], #32
+# CHECK: st1 { v0.2d, v1.2d, v2.2d }, [x0], #48
+# CHECK: st1 { v0.8b, v1.8b, v2.8b, v3.8b }, [x0], x3
0xef,0x75,0x82,0x4c
0xff,0xab,0x9f,0x4c
0x00,0x6c,0x9f,0x4c
0x00,0x20,0x83,0x0c
-# CHECK: st2 {v0.16b, v1.16b}, [x0], x1
-# CHECK: st3 {v15.8h, v16.8h, v17.8h}, [x15], x2
-# CHECK: st4 {v31.4s, v0.4s, v1.4s, v2.4s}, [sp], #64
+# CHECK: st2 { v0.16b, v1.16b }, [x0], x1
+# CHECK: st3 { v15.8h, v16.8h, v17.8h }, [x15], x2
+# CHECK: st4 { v31.4s, v0.4s, v1.4s, v2.4s }, [sp], #64
0x00,0x80,0x81,0x4c
0xef,0x45,0x82,0x4c
0xff,0x0b,0x9f,0x4c
+#----------------------------------------------------------------------
+# Vector load single N-element structure to all lane of N
+# consecutive registers (N = 1,2,3,4)
+#----------------------------------------------------------------------
+# CHECK: ld1r { v0.16b }, [x0]
+# CHECK: ld1r { v15.8h }, [x15]
+# CHECK: ld2r { v31.4s, v0.4s }, [sp]
+# CHECK: ld2r { v0.2d, v1.2d }, [x0]
+# CHECK: ld3r { v0.8b, v1.8b, v2.8b }, [x0]
+# CHECK: ld3r { v15.4h, v16.4h, v17.4h }, [x15]
+# CHECK: ld4r { v31.2s, v0.2s, v1.2s, v2.2s }, [sp]
+# CHECK: ld4r { v31.1d, v0.1d, v1.1d, v2.1d }, [sp]
+0x00,0xc0,0x40,0x4d
+0xef,0xc5,0x40,0x4d
+0xff,0xcb,0x60,0x4d
+0x00,0xcc,0x60,0x4d
+0x00,0xe0,0x40,0x0d
+0xef,0xe5,0x40,0x0d
+0xff,0xeb,0x60,0x0d
+0xff,0xef,0x60,0x0d
+
+#----------------------------------------------------------------------
+# Vector load/store single N-element structure to/from one lane of N
+# consecutive registers (N = 1,2,3,4)
+#----------------------------------------------------------------------
+# CHECK: ld1 { v0.b }[9], [x0]
+# CHECK: ld2 { v15.h, v16.h }[7], [x15]
+# CHECK: ld3 { v31.s, v0.s, v1.s }[3], [sp]
+# CHECK: ld4 { v0.d, v1.d, v2.d, v3.d }[1], [x0]
+# CHECK: st1 { v0.d }[1], [x0]
+# CHECK: st2 { v31.s, v0.s }[3], [sp]
+# CHECK: st3 { v15.h, v16.h, v17.h }[7], [x15]
+# CHECK: st4 { v0.b, v1.b, v2.b, v3.b }[9], [x0]
+0x00,0x04,0x40,0x4d
+0xef,0x59,0x60,0x4d
+0xff,0xb3,0x40,0x4d
+0x00,0xa4,0x60,0x4d
+0x00,0x84,0x00,0x4d
+0xff,0x93,0x20,0x4d
+0xef,0x79,0x00,0x4d
+0x00,0x24,0x20,0x4d
+
+#----------------------------------------------------------------------
+# Post-index of vector load single N-element structure to all lane of N
+# consecutive registers (N = 1,2,3,4)
+#----------------------------------------------------------------------
+# CHECK: ld1r { v0.16b }, [x0], #1
+# CHECK: ld1r { v15.8h }, [x15], #2
+# CHECK: ld2r { v31.4s, v0.4s }, [sp], #8
+# CHECK: ld2r { v0.2d, v1.2d }, [x0], #16
+# CHECK: ld3r { v0.8b, v1.8b, v2.8b }, [x0], #3
+# CHECK: ld3r { v15.4h, v16.4h, v17.4h }, [x15], #6
+# CHECK: ld4r { v31.2s, v0.2s, v1.2s, v2.2s }, [sp], x30
+# CHECK: ld4r { v31.1d, v0.1d, v1.1d, v2.1d }, [sp], x7
+0x00,0xc0,0xdf,0x4d
+0xef,0xc5,0xdf,0x4d
+0xff,0xcb,0xff,0x4d
+0x00,0xcc,0xff,0x4d
+0x00,0xe0,0xdf,0x0d
+0xef,0xe5,0xdf,0x0d
+0xff,0xeb,0xfe,0x0d
+0xff,0xef,0xe7,0x0d
+
+#----------------------------------------------------------------------
+# Post-index of vector load/store single N-element structure to/from
+# one lane of N consecutive registers (N = 1,2,3,4)
+#----------------------------------------------------------------------
+# CHECK: ld1 { v0.b }[9], [x0], #1
+# CHECK: ld2 { v15.h, v16.h }[7], [x15], #4
+# CHECK: ld3 { v31.s, v0.s, v1.s }[3], [sp], x3
+# CHECK: ld4 { v0.d, v1.d, v2.d, v3.d }[1], [x0], #32
+# CHECK: ld4 { v0.h, v1.h, v2.h, v3.h }[7], [x0], x0
+# CHECK: st1 { v0.d }[1], [x0], #8
+# CHECK: st2 { v31.s, v0.s }[3], [sp], #8
+# CHECK: st3 { v15.h, v16.h, v17.h }[7], [x15], #6
+# CHECK: st4 { v0.b, v1.b, v2.b, v3.b }[9], [x0], x5
+0x00,0x04,0xdf,0x4d
+0xef,0x59,0xff,0x4d
+0xff,0xb3,0xc3,0x4d
+0x00,0xa4,0xff,0x4d
+0x00,0x78,0xe0,0x4d
+0x00,0x84,0x9f,0x4d
+0xff,0x93,0xbf,0x4d
+0xef,0x79,0x9f,0x4d
+0x00,0x24,0xa5,0x4d
+
#----------------------------------------------------------------------
# Bitwise extract
#----------------------------------------------------------------------
0x20,0x18,0x02,0x2e
0x20,0x18,0x02,0x6e
-# CHECK: ext v0.8b, v1.8b, v2.8b, #0x3
-# CHECK: ext v0.16b, v1.16b, v2.16b, #0x3
+# CHECK: ext v0.8b, v1.8b, v2.8b, #{{0x3|3}}
+# CHECK: ext v0.16b, v1.16b, v2.16b, #{{0x3|3}}
#----------------------------------------------------------------------
# unzip with 3 same vectors to get primary result
0x28,0x78,0x82,0x0e
0x19,0x78,0x82,0x4e
0x0a,0x78,0xc2,0x4e
+
+#----------------------------------------------------------------------
+# Scalar Floating Point multiply (scalar, by element)
+#----------------------------------------------------------------------
+# CHECK: fmul s0, s1, v1.s[0]
+# CHECK: fmul s0, s1, v1.s[3]
+# CHECK: fmul d0, d1, v1.d[0]
+# CHECK: fmul d0, d1, v1.d[1]
+# CHECK: fmul d15, d15, v15.d[1]
+0x20 0x90 0x81 0x5f
+0x20 0x98 0xa1 0x5f
+0x20 0x90 0xc1 0x5f
+0x20 0x98 0xc1 0x5f
+0xef 0x99 0xcf 0x5f
+
+#----------------------------------------------------------------------
+# Scalar Floating Point multiply extended (scalar, by element)
+#----------------------------------------------------------------------
+# CHECK: fmulx s3, s5, v7.s[0]
+# CHECK: fmulx s3, s5, v7.s[3]
+# CHECK: fmulx s3, s5, v15.s[3]
+# CHECK: fmulx d0, d4, v8.d[0]
+# CHECK: fmulx d0, d4, v8.d[1]
+0xa3 0x90 0x87 0x7f
+0xa3 0x98 0xa7 0x7f
+0xa3 0x98 0xaf 0x7f
+0x80 0x90 0xc8 0x7f
+0x80 0x98 0xc8 0x7f
+
+#----------------------------------------------------------------------
+# Scalar Floating Point fused multiply-add (scalar, by element)
+#----------------------------------------------------------------------
+# CHECK: fmla s0, s1, v1.s[0]
+# CHECK: fmla s0, s1, v1.s[3]
+# CHECK: fmla d0, d1, v1.d[0]
+# CHECK: fmla d0, d1, v1.d[1]
+# CHECK: fmla d15, d15, v15.d[1]
+0x20 0x10 0x81 0x5f
+0x20 0x18 0xa1 0x5f
+0x20 0x10 0xc1 0x5f
+0x20 0x18 0xc1 0x5f
+0xef 0x19 0xcf 0x5f
+
+#----------------------------------------------------------------------
+# Scalar Floating Point fused multiply-sub (scalar, by element)
+#----------------------------------------------------------------------
+# CHECK: fmls s3, s5, v7.s[0]
+# CHECK: fmls s3, s5, v7.s[3]
+# CHECK: fmls s3, s5, v15.s[3]
+# CHECK: fmls d0, d4, v8.d[0]
+# CHECK: fmls d0, d4, v8.d[1]
+0xa3 0x50 0x87 0x5f
+0xa3 0x58 0xa7 0x5f
+0xa3 0x58 0xaf 0x5f
+0x80 0x50 0xc8 0x5f
+0x80 0x58 0xc8 0x5f
+
+#----------------------------------------------------------------------
+# Scalar Signed saturating doubling
+# multiply-add long (scalar, by element)
+#----------------------------------------------------------------------
+# CHECK: sqdmlal s0, h0, v0.h[0]
+# CHECK: sqdmlal s0, h0, v0.h[1]
+# CHECK: sqdmlal s0, h0, v0.h[2]
+# CHECK: sqdmlal s0, h0, v0.h[3]
+# CHECK: sqdmlal s0, h0, v0.h[4]
+# CHECK: sqdmlal s0, h0, v0.h[5]
+# CHECK: sqdmlal s0, h0, v0.h[6]
+# CHECK: sqdmlal s0, h0, v0.h[7]
+# CHECK: sqdmlal d8, s9, v15.s[0]
+# CHECK: sqdmlal d8, s9, v15.s[1]
+# CHECK: sqdmlal d8, s9, v15.s[2]
+# CHECK: sqdmlal d8, s9, v15.s[3]
+0x00 0x30 0x40 0x5f
+0x00 0x30 0x50 0x5f
+0x00 0x30 0x60 0x5f
+0x00 0x30 0x70 0x5f
+0x00 0x38 0x40 0x5f
+0x00 0x38 0x50 0x5f
+0x00 0x38 0x60 0x5f
+0x00 0x38 0x70 0x5f
+0x28 0x31 0x8f 0x5f
+0x28 0x31 0xaf 0x5f
+0x28 0x39 0x8f 0x5f
+0x28 0x39 0xaf 0x5f
+
+#----------------------------------------------------------------------
+# Scalar Signed saturating doubling
+# multiply-sub long (scalar, by element)
+#----------------------------------------------------------------------
+# CHECK: sqdmlsl s0, h0, v0.h[0]
+# CHECK: sqdmlsl s0, h0, v0.h[1]
+# CHECK: sqdmlsl s0, h0, v0.h[2]
+# CHECK: sqdmlsl s0, h0, v0.h[3]
+# CHECK: sqdmlsl s0, h0, v0.h[4]
+# CHECK: sqdmlsl s0, h0, v0.h[5]
+# CHECK: sqdmlsl s0, h0, v0.h[6]
+# CHECK: sqdmlsl s0, h0, v0.h[7]
+# CHECK: sqdmlsl d8, s9, v15.s[0]
+# CHECK: sqdmlsl d8, s9, v15.s[1]
+# CHECK: sqdmlsl d8, s9, v15.s[2]
+# CHECK: sqdmlsl d8, s9, v15.s[3]
+0x00 0x70 0x40 0x5f
+0x00 0x70 0x50 0x5f
+0x00 0x70 0x60 0x5f
+0x00 0x70 0x70 0x5f
+0x00 0x78 0x40 0x5f
+0x00 0x78 0x50 0x5f
+0x00 0x78 0x60 0x5f
+0x00 0x78 0x70 0x5f
+0x28 0x71 0x8f 0x5f
+0x28 0x71 0xaf 0x5f
+0x28 0x79 0x8f 0x5f
+0x28 0x79 0xaf 0x5f
+
+#----------------------------------------------------------------------
+# Scalar Signed saturating doubling multiply long (scalar, by element)
+#----------------------------------------------------------------------
+# CHECK: sqdmull s1, h1, v1.h[0]
+# CHECK: sqdmull s1, h1, v1.h[1]
+# CHECK: sqdmull s1, h1, v1.h[2]
+# CHECK: sqdmull s1, h1, v1.h[3]
+# CHECK: sqdmull s1, h1, v1.h[4]
+# CHECK: sqdmull s1, h1, v1.h[5]
+# CHECK: sqdmull s1, h1, v1.h[6]
+# CHECK: sqdmull s1, h1, v1.h[7]
+# CHECK: sqdmull d1, s1, v4.s[0]
+# CHECK: sqdmull d1, s1, v4.s[1]
+# CHECK: sqdmull d1, s1, v4.s[2]
+# CHECK: sqdmull d1, s1, v4.s[3]
+0x21 0xb0 0x41 0x5f
+0x21 0xb0 0x51 0x5f
+0x21 0xb0 0x61 0x5f
+0x21 0xb0 0x71 0x5f
+0x21 0xb8 0x41 0x5f
+0x21 0xb8 0x51 0x5f
+0x21 0xb8 0x61 0x5f
+0x21 0xb8 0x71 0x5f
+0x21 0xb0 0x84 0x5f
+0x21 0xb0 0xa4 0x5f
+0x21 0xb8 0x84 0x5f
+0x21 0xb8 0xa4 0x5f
+
+#----------------------------------------------------------------------
+# Scalar Signed saturating doubling multiply returning
+# high half (scalar, by element)
+#----------------------------------------------------------------------
+# CHECK: sqdmulh h7, h1, v14.h[0]
+# CHECK: sqdmulh h7, h15, v8.h[1]
+# CHECK: sqdmulh h7, h15, v8.h[2]
+# CHECK: sqdmulh h7, h15, v8.h[3]
+# CHECK: sqdmulh h7, h15, v8.h[4]
+# CHECK: sqdmulh h7, h15, v8.h[5]
+# CHECK: sqdmulh h7, h15, v8.h[6]
+# CHECK: sqdmulh h7, h15, v8.h[7]
+# CHECK: sqdmulh s15, s3, v4.s[0]
+# CHECK: sqdmulh s15, s14, v16.s[1]
+# CHECK: sqdmulh s15, s15, v16.s[2]
+# CHECK: sqdmulh s15, s16, v17.s[3]
+0x27 0xc0 0x4e 0x5f
+0xe7 0xc1 0x58 0x5f
+0xe7 0xc1 0x68 0x5f
+0xe7 0xc1 0x78 0x5f
+0xe7 0xc9 0x48 0x5f
+0xe7 0xc9 0x58 0x5f
+0xe7 0xc9 0x68 0x5f
+0xe7 0xc9 0x78 0x5f
+0x6f 0xc0 0x84 0x5f
+0xcf 0xc1 0xb0 0x5f
+0xef 0xc9 0x90 0x5f
+0x0f 0xca 0xb1 0x5f
+
+#----------------------------------------------------------------------
+# Scalar Signed saturating rounding doubling multiply
+# returning high half (scalar, by element)
+#----------------------------------------------------------------------
+# CHECK: sqrdmulh h7, h1, v14.h[0]
+# CHECK: sqrdmulh h7, h15, v8.h[1]
+# CHECK: sqrdmulh h7, h15, v8.h[2]
+# CHECK: sqrdmulh h7, h15, v8.h[3]
+# CHECK: sqrdmulh h7, h15, v8.h[4]
+# CHECK: sqrdmulh h7, h15, v8.h[5]
+# CHECK: sqrdmulh h7, h15, v8.h[6]
+# CHECK: sqrdmulh h7, h15, v8.h[7]
+# CHECK: sqrdmulh s15, s3, v4.s[0]
+# CHECK: sqrdmulh s15, s14, v16.s[1]
+# CHECK: sqrdmulh s15, s15, v16.s[2]
+# CHECK: sqrdmulh s15, s16, v17.s[3]
+0x27 0xd0 0x4e 0x5f
+0xe7 0xd1 0x58 0x5f
+0xe7 0xd1 0x68 0x5f
+0xe7 0xd1 0x78 0x5f
+0xe7 0xd9 0x48 0x5f
+0xe7 0xd9 0x58 0x5f
+0xe7 0xd9 0x68 0x5f
+0xe7 0xd9 0x78 0x5f
+0x6f 0xd0 0x84 0x5f
+0xcf 0xd1 0xb0 0x5f
+0xef 0xd9 0x90 0x5f
+0x0f 0xda 0xb1 0x5f
+
+#----------------------------------------------------------------------
+#Duplicate element (scalar)
+#----------------------------------------------------------------------
+# CHECK: {{dup|mov}} b0, v0.b[15]
+# CHECK: {{dup|mov}} h2, v31.h[5]
+# CHECK: {{dup|mov}} s17, v2.s[2]
+# CHECK: {{dup|mov}} d6, v12.d[1]
+0x00 0x04 0x1f 0x5e
+0xe2 0x07 0x16 0x5e
+0x51 0x04 0x14 0x5e
+0x86 0x05 0x18 0x5e
+
+#----------------------------------------------------------------------
+# Table look up
+#----------------------------------------------------------------------
+0x20,0x00,0x02,0x0e
+0xf0,0x23,0x02,0x0e
+0x20,0x40,0x02,0x0e
+0xf0,0x62,0x02,0x0e
+# CHECK: tbl v0.8b, { v1.16b }, v2.8b
+# CHECK: tbl v16.8b, { v31.16b, v0.16b }, v2.8b
+# CHECK: tbl v0.8b, { v1.16b, v2.16b, v3.16b }, v2.8b
+# CHECK: tbl v16.8b, { v23.16b, v24.16b, v25.16b, v26.16b }, v2.8b
+
+0x20,0x00,0x02,0x4e
+0xf0,0x23,0x02,0x4e
+0x20,0x40,0x02,0x4e
+0xe0,0x63,0x02,0x4e
+# CHECK: tbl v0.16b, { v1.16b }, v2.16b
+# CHECK: tbl v16.16b, { v31.16b, v0.16b }, v2.16b
+# CHECK: tbl v0.16b, { v1.16b, v2.16b, v3.16b }, v2.16b
+# CHECK: tbl v0.16b, { v31.16b, v0.16b, v1.16b, v2.16b }, v2.16b
+
+0x20,0x10,0x02,0x0e
+0xf0,0x33,0x02,0x0e
+0x20,0x50,0x02,0x0e
+0xf0,0x72,0x02,0x0e
+# CHECK: tbx v0.8b, { v1.16b }, v2.8b
+# CHECK: tbx v16.8b, { v31.16b, v0.16b }, v2.8b
+# CHECK: tbx v0.8b, { v1.16b, v2.16b, v3.16b }, v2.8b
+# CHECK: tbx v16.8b, { v23.16b, v24.16b, v25.16b, v26.16b }, v2.8b
+
+0x20,0x10,0x02,0x4e
+0xf0,0x33,0x02,0x4e
+0x20,0x50,0x02,0x4e
+0xf0,0x73,0x02,0x4e
+# CHECK: tbx v0.16b, { v1.16b }, v2.16b
+# CHECK: tbx v16.16b, { v31.16b, v0.16b }, v2.16b
+# CHECK: tbx v0.16b, { v1.16b, v2.16b, v3.16b }, v2.16b
+# CHECK: tbx v16.16b, { v31.16b, v0.16b, v1.16b, v2.16b }, v2.16b
+
+#----------------------------------------------------------------------
+# Scalar Floating-point Convert To Lower Precision Narrow, Rounding To
+# Odd
+#----------------------------------------------------------------------
+# CHECK: fcvtxn s22, d13
+0xb6,0x69,0x61,0x7e
+
+#----------------------------------------------------------------------
+# Scalar Floating-point Convert To Signed Integer, Rounding To Nearest
+# With Ties To Away
+#----------------------------------------------------------------------
+# CHECK: fcvtas s12, s13
+# CHECK: fcvtas d21, d14
+
+0xac,0xc9,0x21,0x5e
+0xd5,0xc9,0x61,0x5e
+
+#----------------------------------------------------------------------
+# Scalar Floating-point Convert To Unsigned Integer, Rounding To
+# Nearest With Ties To Away
+#----------------------------------------------------------------------
+# CHECK: fcvtau s12, s13
+# CHECK: fcvtau d21, d14
+0xac,0xc9,0x21,0x7e
+0xd5,0xc9,0x61,0x7e
+
+#----------------------------------------------------------------------
+# Scalar Floating-point Convert To Signed Integer, Rounding Toward
+# Minus Infinity
+#----------------------------------------------------------------------
+# CHECK: fcvtms s22, s13
+# CHECK: fcvtms d21, d14
+0xb6,0xb9,0x21,0x5e
+0xd5,0xb9,0x61,0x5e
+
+#----------------------------------------------------------------------
+# Scalar Floating-point Convert To Unsigned Integer, Rounding Toward
+# Minus Infinity
+#----------------------------------------------------------------------
+# CHECK: fcvtmu s12, s13
+# CHECK: fcvtmu d21, d14
+0xac,0xb9,0x21,0x7e
+0xd5,0xb9,0x61,0x7e
+
+#----------------------------------------------------------------------
+# Scalar Floating-point Convert To Signed Integer, Rounding To Nearest
+# With Ties To Even
+#----------------------------------------------------------------------
+
+# CHECK: fcvtns s22, s13
+# CHECK: fcvtns d21, d14
+
+0xb6,0xa9,0x21,0x5e
+0xd5,0xa9,0x61,0x5e
+
+#----------------------------------------------------------------------
+# Scalar Floating-point Convert To Unsigned Integer, Rounding To
+# Nearest With Ties To Even
+#----------------------------------------------------------------------
+
+# CHECK: fcvtnu s12, s13
+# CHECK: fcvtnu d21, d14
+0xac,0xa9,0x21,0x7e
+0xd5,0xa9,0x61,0x7e
+
+#----------------------------------------------------------------------
+# Scalar Floating-point Convert To Signed Integer, Rounding Toward
+# Positive Infinity
+#----------------------------------------------------------------------
+# CHECK: fcvtps s22, s13
+# CHECK: fcvtps d21, d14
+0xb6,0xa9,0xa1,0x5e
+0xd5,0xa9,0xe1,0x5e
+
+#----------------------------------------------------------------------
+# Scalar Floating-point Convert To Unsigned Integer, Rounding Toward
+# Positive Infinity
+#----------------------------------------------------------------------
+# CHECK: fcvtpu s12, s13
+# CHECK: fcvtpu d21, d14
+0xac,0xa9,0xa1,0x7e
+0xd5,0xa9,0xe1,0x7e
+
+#----------------------------------------------------------------------
+# Scalar Floating-point Convert To Signed Integer, Rounding Toward Zero
+#----------------------------------------------------------------------
+# CHECK: fcvtzs s12, s13
+# CHECK: fcvtzs d21, d14
+0xac,0xb9,0xa1,0x5e
+0xd5,0xb9,0xe1,0x5e
+
+#----------------------------------------------------------------------
+# Scalar Floating-point Convert To Unsigned Integer, Rounding Toward
+# Zero
+#----------------------------------------------------------------------
+# CHECK: fcvtzu s12, s13
+# CHECK: fcvtzu d21, d14
+0xac,0xb9,0xa1,0x7e
+0xd5,0xb9,0xe1,0x7e
+
+#----------------------------------------------------------------------
+# Scalar Floating-point Absolute Difference
+#----------------------------------------------------------------------
+# CHECK: fabd s29, s24, s20
+# CHECK: fabd d29, d24, d20
+0x1d,0xd7,0xb4,0x7e
+0x1d,0xd7,0xf4,0x7e