[ARM] Add MVE vector shift instructions.

[simon_tatham]

Summary

This includes saturating and non-saturating shifts, both with
immediate shift count and with the shift counts given by another
vector register; VSHLC (in which the bits shifted out of each active
vector lane are shifted in to the next active lane); and also VMOVL,
which is enough like an immediate shift that it didn't fit too badly
in this category.

[ostannard]

llvm/lib/Target/ARM/ARMInstrMVE.td

785 ↗

(On Diff #202207)

Why are the R and Q registers in the opposite order in the input and output operands?

815 ↗

(On Diff #202207)

Some of these names begin with "t2" (matching the other Thumb2 instructions), and some don't. Unless there's a pattern which I've missed, could pick one naming convention and stick to it?

909 ↗

(On Diff #202207)

These ones seem to be be using the t1/t2 prefix to distinguish between different 32-bit encodings, rather than between the 16-bit and 32-bit encodings from the reference manual. I'd prefer if we avoided using the encoding numbers at all, because these don't tend to be stable between versions of the architecture.

1044 ↗

(On Diff #202207)

This class is only used once, can it be merged into MVE_shift_helper?

llvm/lib/Target/ARM/AsmParser/ARMAsmParser.cpp

5984 ↗

(On Diff #202207)

What are these special cases for? Could this be moved into isMnemonicVPTPredicable?

6610 ↗

(On Diff #202207)

Does this need to be an exact equality check, not just startswith? Otherwise, I think this will transform "vmov<whatever>lt" into just "vmovlt".

llvm/lib/Target/ARM/Disassembler/ARMDisassembler.cpp

6089 ↗

(On Diff #202207)

Is this not already handled by the table-generated disassembler?

[simon_tatham]

Remastered patch to apply cleanly against current trunk.

[simon_tatham]

llvm/lib/Target/ARM/ARMInstrMVE.td

785 ↗

(On Diff #202207)

I think this one was intentional, because it reflects the logical order of the operand pairs. VSHLC shifts a Q-register left, filling in the rightmost bits from a GPR, and as output it produces a GPR full of the bits shifted off from the left. So you start with Qx,Ry which logically make up a single bit sequence if arranged in that order, and you end up with Ry,Qx which now make up a shifted version of the same bit sequence if arranged in that order.

815 ↗

(On Diff #202207)

I suspect this was just a case of 'add a prefix wherever it's necessary to prevent a clash with an existing instruction'.

Some MVE instructions have already been committed, so I've opened D63492 to make those more consistent. Also revised D62671 in line with this comment. I'll try to arrange that all the MVE instructions have names beginning MVE_.

llvm/lib/Target/ARM/AsmParser/ARMAsmParser.cpp

5984 ↗

(On Diff #202207)

I'm a bit unclear on this, but I think the problem here is that isMnemonicVPTPredicable has to be able to answer questions about the version of the mnemonic with a T or E suffix (because it's called by SplitMnemonic) and also about the version without one (because it's called by getMnemonicAcceptInfo after the split).

On that basis, these special cases are all things that isMnemonicVPTPredicable properly ought to return true for, either because they are things on which a T/E suffix would be legal (e.g. vqshrnt, in which the existing t means 'top half' for a narrowing instruction), or because they actually have one (vmovlt, unless that's a scalar vmov with an ordinary lt conditional suffix).

If we broke up isMnemonicVPTPredicable into two separate functions, with semantics "is this a mnemonic that might already include a T/E suffix?" and "is this a mnemonic that it would be legal to put a T/E suffix on the end of?", then we could probably get rid of most of these special cases ... (a) provided I haven't completely misunderstood all of this; (b) at the cost of making isMnemonicVPTPredicable itself hairier by more than we save here; (c) and it wouldn't even get rid of all of them, because I think the reason vmovlt is treated unusually here is to handle the lexing ambiguity between vmovl t and vmov lt.

llvm/lib/Target/ARM/Disassembler/ARMDisassembler.cpp

6089 ↗

(On Diff #202207)

I think this custom decoder was written because of the combined size + shift count field in the encoding for these instructions. But I've found an alternative way to express it in pure Tablegen, so yes, it looks as if we can get rid of this.

[simon_tatham]

Renamed all instructions for consistency; removed the unnecessary intermediate class; reworked VSHLL to manage without a custom C++ decoder function.

[ostannard]

LGTM

llvm/trunk/lib/Target/ARM/ARMInstrInfo.td

 def imm1_16 : Operand<i32>, ImmLeaf<i32, [{
     return Imm > 0 && Imm <= 16;
   }],
     imm1_16_XFORM> {
   let PrintMethod = "printImmPlusOneOperand";
   let ParserMatchClass = Imm1_16AsmOperand;
 }
 
+def MVEShiftImm1_7AsmOperand: ImmAsmOperand<1,7> {
+  let Name = "MVEShiftImm1_7";
+  // Reason we're doing this is because instruction vshll.s8 t1 encoding
+  // accepts 1,7 but the t2 encoding accepts 8.  By doing this we can get a
+  // better diagnostic message if someone uses bigger immediate than the t1/t2
+  // encodings allow.
+  let DiagnosticString = "operand must be an immediate in the range [1,8]";
+}
+def mve_shift_imm1_7 : Operand<i32> {
+  let ParserMatchClass = MVEShiftImm1_7AsmOperand;
+  let EncoderMethod = "getMVEShiftImmOpValue";
+}
+
+def MVEShiftImm1_15AsmOperand: ImmAsmOperand<1,15> {
+  let Name = "MVEShiftImm1_15";
+  // Reason we're doing this is because instruction vshll.s16 t1 encoding
+  // accepts 1,15 but the t2 encoding accepts 16.  By doing this we can get a
+  // better diagnostic message if someone uses bigger immediate than the t1/t2
+  // encodings allow.
+  let DiagnosticString = "operand must be an immediate in the range [1,16]";
+}
+def mve_shift_imm1_15 : Operand<i32> {
+  let ParserMatchClass = MVEShiftImm1_15AsmOperand;
+  let EncoderMethod = "getMVEShiftImmOpValue";
+}
+
 // Define ARM specific addressing modes.
 // addrmode_imm12 := reg +/- imm12
 //
 def MemImm12OffsetAsmOperand : AsmOperandClass { let Name = "MemImm12Offset"; }
 class AddrMode_Imm12 : MemOperand,
                      ComplexPattern<i32, 2, "SelectAddrModeImm12", []> {
   // 12-bit immediate operand. Note that instructions using this encode
   // #0 and #-0 differently. We flag #-0 as the magic value INT32_MIN. All other

llvm/trunk/lib/Target/ARM/ARMInstrMVE.td

multiclass MVE_VMINMAX_all_sizes<string iname, bit bit_4> {
   def u32 : MVE_VMINMAX<iname, "u32", 0b1, 0b10, bit_4>;
 }
 
 defm MVE_VMAX : MVE_VMINMAX_all_sizes<"vmax", 0b0>;
 defm MVE_VMIN : MVE_VMINMAX_all_sizes<"vmin", 0b1>;
 
 // end of mve_comp instructions
 
+// start of mve_imm_shift instructions
+
+def MVE_VSHLC : MVE_p<(outs rGPR:$RdmDest, MQPR:$Qd),
+                      (ins MQPR:$QdSrc, rGPR:$RdmSrc, long_shift:$imm),
+                      NoItinerary, "vshlc", "", "$QdSrc, $RdmSrc, $imm",
+                      vpred_n, "$RdmDest = $RdmSrc,$Qd = $QdSrc"> {
+  bits<5> imm;
+  bits<4> Qd;
+  bits<4> RdmDest;
+
+  let Inst{28} = 0b0;
+  let Inst{25-23} = 0b101;
+  let Inst{22} = Qd{3};
+  let Inst{21} = 0b1;
+  let Inst{20-16} = imm{4-0};
+  let Inst{15-13} = Qd{2-0};
+  let Inst{12-4} = 0b011111100;
+  let Inst{3-0} = RdmDest{3-0};
+}
+
+class MVE_shift_imm<dag oops, dag iops, string iname, string suffix,
+                    string ops, vpred_ops vpred, string cstr,
+                    list<dag> pattern=[]>
+  : MVE_p<oops, iops, NoItinerary, iname, suffix, ops, vpred, cstr, pattern> {
+  bits<4> Qd;
+  bits<4> Qm;
+
+  let Inst{22} = Qd{3};
+  let Inst{15-13} = Qd{2-0};
+  let Inst{5} = Qm{3};
+  let Inst{3-1} = Qm{2-0};
+}
+
+class MVE_VMOVL<string iname, string suffix, bits<2> sz, bit U,
+              list<dag> pattern=[]>
+  : MVE_shift_imm<(outs MQPR:$Qd), (ins MQPR:$Qm),
+                  iname, suffix, "$Qd, $Qm", vpred_r, "",
+                  pattern> {
+  let Inst{28} = U;
+  let Inst{25-23} = 0b101;
+  let Inst{21} = 0b1;
+  let Inst{20-19} = sz{1-0};
+  let Inst{18-16} = 0b000;
+  let Inst{11-6} = 0b111101;
+  let Inst{4} = 0b0;
+  let Inst{0} = 0b0;
+}
+
+multiclass MVE_VMOVL_shift_half<string iname, string suffix, bits<2> sz, bit U,
+                                list<dag> pattern=[]> {
+  def bh : MVE_VMOVL<!strconcat(iname, "b"), suffix, sz, U, pattern> {
+    let Inst{12} = 0b0;
+  }
+  def th : MVE_VMOVL<!strconcat(iname, "t"), suffix, sz, U, pattern> {
+    let Inst{12} = 0b1;
+  }
+}
+
+defm MVE_VMOVLs8 : MVE_VMOVL_shift_half<"vmovl", "s8", 0b01, 0b0>;
+defm MVE_VMOVLu8 : MVE_VMOVL_shift_half<"vmovl", "u8", 0b01, 0b1>;
+defm MVE_VMOVLs16 : MVE_VMOVL_shift_half<"vmovl", "s16", 0b10, 0b0>;
+defm MVE_VMOVLu16 : MVE_VMOVL_shift_half<"vmovl", "u16", 0b10, 0b1>;
+
+class MVE_VSHLL_imm<string iname, string suffix, bit U, bit th,
+                    dag immops, list<dag> pattern=[]>
+  : MVE_shift_imm<(outs MQPR:$Qd), !con((ins MQPR:$Qm), immops),
+                  iname, suffix, "$Qd, $Qm, $imm", vpred_r, "", pattern> {
+  let Inst{28} = U;
+  let Inst{25-23} = 0b101;
+  let Inst{21} = 0b1;
+  let Inst{12} = th;
+  let Inst{11-6} = 0b111101;
+  let Inst{4} = 0b0;
+  let Inst{0} = 0b0;
+}
+
+// The immediate VSHLL instructions accept shift counts from 1 up to
+// the lane width (8 or 16), but the full-width shifts have an
+// entirely separate encoding, given below with 'lw' in the name.
+
+class MVE_VSHLL_imm8<string iname, string suffix,
+                     bit U, bit th, list<dag> pattern=[]>
+  : MVE_VSHLL_imm<iname, suffix, U, th, (ins mve_shift_imm1_7:$imm), pattern> {
+  bits<3> imm;
+  let Inst{20-19} = 0b01;
+  let Inst{18-16} = imm;
+}
+
+class MVE_VSHLL_imm16<string iname, string suffix,
+                      bit U, bit th, list<dag> pattern=[]>
+  : MVE_VSHLL_imm<iname, suffix, U, th, (ins mve_shift_imm1_15:$imm), pattern> {
+  bits<4> imm;
+  let Inst{20} = 0b1;
+  let Inst{19-16} = imm;
+}
+
+def MVE_VSHLL_imms8bh  : MVE_VSHLL_imm8 <"vshllb", "s8", 0b0, 0b0>;
+def MVE_VSHLL_imms8th  : MVE_VSHLL_imm8 <"vshllt", "s8", 0b0, 0b1>;
+def MVE_VSHLL_immu8bh  : MVE_VSHLL_imm8 <"vshllb", "u8", 0b1, 0b0>;
+def MVE_VSHLL_immu8th  : MVE_VSHLL_imm8 <"vshllt", "u8", 0b1, 0b1>;
+def MVE_VSHLL_imms16bh : MVE_VSHLL_imm16<"vshllb", "s16", 0b0, 0b0>;
+def MVE_VSHLL_imms16th : MVE_VSHLL_imm16<"vshllt", "s16", 0b0, 0b1>;
+def MVE_VSHLL_immu16bh : MVE_VSHLL_imm16<"vshllb", "u16", 0b1, 0b0>;
+def MVE_VSHLL_immu16th : MVE_VSHLL_imm16<"vshllt", "u16", 0b1, 0b1>;
+
+class MVE_VSHLL_by_lane_width<string iname, string suffix, bits<2> size,
+                              bit U, string ops, list<dag> pattern=[]>
+  : MVE_shift_imm<(outs MQPR:$Qd), (ins MQPR:$Qm),
+                  iname, suffix, ops, vpred_r, "", pattern> {
+  let Inst{28} = U;
+  let Inst{25-23} = 0b100;
+  let Inst{21-20} = 0b11;
+  let Inst{19-18} = size{1-0};
+  let Inst{17-16} = 0b01;
+  let Inst{11-6} = 0b111000;
+  let Inst{4} = 0b0;
+  let Inst{0} = 0b1;
+}
+
+multiclass MVE_VSHLL_lw<string iname, string suffix, bits<2> sz, bit U,
+                              string ops, list<dag> pattern=[]> {
+  def bh : MVE_VSHLL_by_lane_width<iname#"b", suffix, sz, U, ops, pattern> {
+    let Inst{12} = 0b0;
+  }
+  def th : MVE_VSHLL_by_lane_width<iname#"t", suffix, sz, U, ops, pattern> {
+    let Inst{12} = 0b1;
+  }
+}
+
+defm MVE_VSHLL_lws8  : MVE_VSHLL_lw<"vshll", "s8",  0b00, 0b0, "$Qd, $Qm, #8">;
+defm MVE_VSHLL_lws16 : MVE_VSHLL_lw<"vshll", "s16", 0b01, 0b0, "$Qd, $Qm, #16">;
+defm MVE_VSHLL_lwu8  : MVE_VSHLL_lw<"vshll", "u8",  0b00, 0b1, "$Qd, $Qm, #8">;
+defm MVE_VSHLL_lwu16 : MVE_VSHLL_lw<"vshll", "u16", 0b01, 0b1, "$Qd, $Qm, #16">;
+
+class MVE_VxSHRN<string iname, string suffix, bit bit_12, bit bit_28,
+               dag immops, list<dag> pattern=[]>
+  : MVE_shift_imm<(outs MQPR:$Qd), !con((ins MQPR:$QdSrc, MQPR:$Qm), immops),
+                  iname, suffix, "$Qd, $Qm, $imm", vpred_n, "$Qd = $QdSrc",
+                  pattern> {
+  bits<5> imm;
+
+  let Inst{28} = bit_28;
+  let Inst{25-23} = 0b101;
+  let Inst{21} = 0b0;
+  let Inst{20-16} = imm{4-0};
+  let Inst{12} = bit_12;
+  let Inst{11-6} = 0b111111;
+  let Inst{4} = 0b0;
+  let Inst{0} = 0b1;
+}
+
+def MVE_VRSHRNi16bh : MVE_VxSHRN<
+    "vrshrnb", "i16", 0b0, 0b1, (ins shr_imm8:$imm)> {
+  let Inst{20-19} = 0b01;
+}
+def MVE_VRSHRNi16th : MVE_VxSHRN<
+    "vrshrnt", "i16", 0b1, 0b1,(ins shr_imm8:$imm)> {
+  let Inst{20-19} = 0b01;
+}
+def MVE_VRSHRNi32bh : MVE_VxSHRN<
+    "vrshrnb", "i32", 0b0, 0b1, (ins shr_imm16:$imm)> {
+  let Inst{20} = 0b1;
+}
+def MVE_VRSHRNi32th : MVE_VxSHRN<
+    "vrshrnt", "i32", 0b1, 0b1, (ins shr_imm16:$imm)> {
+  let Inst{20} = 0b1;
+}
+
+def MVE_VSHRNi16bh : MVE_VxSHRN<
+    "vshrnb", "i16", 0b0, 0b0, (ins shr_imm8:$imm)> {
+  let Inst{20-19} = 0b01;
+}
+def MVE_VSHRNi16th : MVE_VxSHRN<
+    "vshrnt", "i16", 0b1, 0b0, (ins shr_imm8:$imm)> {
+  let Inst{20-19} = 0b01;
+}
+def MVE_VSHRNi32bh : MVE_VxSHRN<
+    "vshrnb", "i32", 0b0, 0b0, (ins shr_imm16:$imm)> {
+  let Inst{20} = 0b1;
+}
+def MVE_VSHRNi32th : MVE_VxSHRN<
+    "vshrnt", "i32", 0b1, 0b0, (ins shr_imm16:$imm)> {
+  let Inst{20} = 0b1;
+}
+
+class MVE_VxQRSHRUN<string iname, string suffix, bit bit_28, bit bit_12, dag immops,
+                 list<dag> pattern=[]>
+  : MVE_shift_imm<(outs MQPR:$Qd), !con((ins MQPR:$QdSrc, MQPR:$Qm), immops),
+                  iname, suffix, "$Qd, $Qm, $imm", vpred_n, "$Qd = $QdSrc",
+                  pattern> {
+  bits<5> imm;
+
+  let Inst{28} = bit_28;
+  let Inst{25-23} = 0b101;
+  let Inst{21} = 0b0;
+  let Inst{20-16} = imm{4-0};
+  let Inst{12} = bit_12;
+  let Inst{11-6} = 0b111111;
+  let Inst{4} = 0b0;
+  let Inst{0} = 0b0;
+}
+
+def MVE_VQRSHRUNs16bh : MVE_VxQRSHRUN<
+    "vqrshrunb", "s16", 0b1, 0b0, (ins shr_imm8:$imm)> {
+  let Inst{20-19} = 0b01;
+}
+def MVE_VQRSHRUNs16th : MVE_VxQRSHRUN<
+    "vqrshrunt", "s16", 0b1, 0b1, (ins shr_imm8:$imm)> {
+  let Inst{20-19} = 0b01;
+}
+def MVE_VQRSHRUNs32bh : MVE_VxQRSHRUN<
+    "vqrshrunb", "s32", 0b1, 0b0, (ins shr_imm16:$imm)> {
+  let Inst{20} = 0b1;
+}
+def MVE_VQRSHRUNs32th : MVE_VxQRSHRUN<
+    "vqrshrunt", "s32", 0b1, 0b1, (ins shr_imm16:$imm)> {
+  let Inst{20} = 0b1;
+}
+
+def MVE_VQSHRUNs16bh : MVE_VxQRSHRUN<
+    "vqshrunb", "s16", 0b0, 0b0, (ins shr_imm8:$imm)> {
+  let Inst{20-19} = 0b01;
+}
+def MVE_VQSHRUNs16th : MVE_VxQRSHRUN<
+    "vqshrunt", "s16", 0b0, 0b1, (ins shr_imm8:$imm)> {
+  let Inst{20-19} = 0b01;
+}
+def MVE_VQSHRUNs32bh : MVE_VxQRSHRUN<
+    "vqshrunb", "s32", 0b0, 0b0, (ins shr_imm16:$imm)> {
+  let Inst{20} = 0b1;
+}
+def MVE_VQSHRUNs32th : MVE_VxQRSHRUN<
+    "vqshrunt", "s32", 0b0, 0b1, (ins shr_imm16:$imm)> {
+  let Inst{20} = 0b1;
+}
+
+class MVE_VxQRSHRN<string iname, string suffix, bit bit_0, bit bit_12,
+                   dag immops, list<dag> pattern=[]>
+  : MVE_shift_imm<(outs MQPR:$Qd), !con((ins MQPR:$QdSrc, MQPR:$Qm), immops),
+                  iname, suffix, "$Qd, $Qm, $imm", vpred_n, "$Qd = $QdSrc",
+                  pattern> {
+  bits<5> imm;
+
+  let Inst{25-23} = 0b101;
+  let Inst{21} = 0b0;
+  let Inst{20-16} = imm{4-0};
+  let Inst{12} = bit_12;
+  let Inst{11-6} = 0b111101;
+  let Inst{4} = 0b0;
+  let Inst{0} = bit_0;
+}
+
+multiclass MVE_VxQRSHRN_types<string iname, bit bit_0, bit bit_12> {
+  def s16 : MVE_VxQRSHRN<iname, "s16", bit_0, bit_12, (ins shr_imm8:$imm)> {
+    let Inst{28} = 0b0;
+    let Inst{20-19} = 0b01;
+  }
+  def u16 : MVE_VxQRSHRN<iname, "u16", bit_0, bit_12, (ins shr_imm8:$imm)> {
+    let Inst{28} = 0b1;
+    let Inst{20-19} = 0b01;
+  }
+  def s32 : MVE_VxQRSHRN<iname, "s32", bit_0, bit_12, (ins shr_imm16:$imm)> {
+    let Inst{28} = 0b0;
+    let Inst{20} = 0b1;
+  }
+  def u32 : MVE_VxQRSHRN<iname, "u32", bit_0, bit_12, (ins shr_imm16:$imm)> {
+    let Inst{28} = 0b1;
+    let Inst{20} = 0b1;
+  }
+}
+
+defm MVE_VQRSHRNbh : MVE_VxQRSHRN_types<"vqrshrnb", 0b1, 0b0>;
+defm MVE_VQRSHRNth : MVE_VxQRSHRN_types<"vqrshrnt", 0b1, 0b1>;
+defm MVE_VQSHRNbh  : MVE_VxQRSHRN_types<"vqshrnb", 0b0, 0b0>;
+defm MVE_VQSHRNth  : MVE_VxQRSHRN_types<"vqshrnt", 0b0, 0b1>;
+
+// end of mve_imm_shift instructions
+
+// start of mve_shift instructions
+
+class MVE_shift_by_vec<string iname, string suffix, bit U,
+                       bits<2> size, bit bit_4, bit bit_8>
+  : MVE_p<(outs MQPR:$Qd), (ins MQPR:$Qm, MQPR:$Qn), NoItinerary,
+           iname, suffix, "$Qd, $Qm, $Qn", vpred_r, "", []> {
+  // Shift instructions which take a vector of shift counts
+  bits<4> Qd;
+  bits<4> Qm;
+  bits<4> Qn;
+
+  let Inst{28} = U;
+  let Inst{25-24} = 0b11;
+  let Inst{23} = 0b0;
+  let Inst{22} = Qd{3};
+  let Inst{21-20} = size;
+  let Inst{19-17} = Qn{2-0};
+  let Inst{16} = 0b0;
+  let Inst{15-13} = Qd{2-0};
+  let Inst{12-9} = 0b0010;
+  let Inst{8} = bit_8;
+  let Inst{7} = Qn{3};
+  let Inst{6} = 0b1;
+  let Inst{5} = Qm{3};
+  let Inst{4} = bit_4;
+  let Inst{3-1} = Qm{2-0};
+  let Inst{0} = 0b0;
+}
+
+multiclass mve_shift_by_vec_multi<string iname, bit bit_4, bit bit_8> {
+  def s8  : MVE_shift_by_vec<iname, "s8", 0b0, 0b00, bit_4, bit_8>;
+  def s16 : MVE_shift_by_vec<iname, "s16", 0b0, 0b01, bit_4, bit_8>;
+  def s32 : MVE_shift_by_vec<iname, "s32", 0b0, 0b10, bit_4, bit_8>;
+  def u8  : MVE_shift_by_vec<iname, "u8", 0b1, 0b00, bit_4, bit_8>;
+  def u16 : MVE_shift_by_vec<iname, "u16", 0b1, 0b01, bit_4, bit_8>;
+  def u32 : MVE_shift_by_vec<iname, "u32", 0b1, 0b10, bit_4, bit_8>;
+}
+
+defm MVE_VSHL_by_vec   : mve_shift_by_vec_multi<"vshl",   0b0, 0b0>;
+defm MVE_VQSHL_by_vec  : mve_shift_by_vec_multi<"vqshl",  0b1, 0b0>;
+defm MVE_VQRSHL_by_vec : mve_shift_by_vec_multi<"vqrshl", 0b1, 0b1>;
+defm MVE_VRSHL_by_vec  : mve_shift_by_vec_multi<"vrshl",  0b0, 0b1>;
+
+class MVE_shift_with_imm<string iname, string suffix, dag oops, dag iops,
+                         string ops, vpred_ops vpred, string cstr,
+                         list<dag> pattern=[]>
+  : MVE_p<oops, iops, NoItinerary, iname, suffix, ops, vpred, cstr, pattern> {
+  bits<4> Qd;
+  bits<4> Qm;
+
+  let Inst{23} = 0b1;
+  let Inst{22} = Qd{3};
+  let Inst{15-13} = Qd{2-0};
+  let Inst{12-11} = 0b00;
+  let Inst{7-6} = 0b01;
+  let Inst{5} = Qm{3};
+  let Inst{4} = 0b1;
+  let Inst{3-1} = Qm{2-0};
+  let Inst{0} = 0b0;
+}
+
+class MVE_VSxI_imm<string iname, string suffix, bit bit_8, dag imm>
+  : MVE_shift_with_imm<iname, suffix, (outs MQPR:$Qd),
+                       !con((ins MQPR:$Qd_src, MQPR:$Qm), imm),
+                       "$Qd, $Qm, $imm", vpred_n, "$Qd = $Qd_src"> {
+  bits<6> imm;
+  let Inst{28} = 0b1;
+  let Inst{25-24} = 0b11;
+  let Inst{21-16} = imm;
+  let Inst{10-9} = 0b10;
+  let Inst{8} = bit_8;
+}
+
+def MVE_VSRIimm8 : MVE_VSxI_imm<"vsri", "8", 0b0, (ins shr_imm8:$imm)> {
+  let Inst{21-19} = 0b001;
+}
+
+def MVE_VSRIimm16 : MVE_VSxI_imm<"vsri", "16", 0b0, (ins shr_imm16:$imm)> {
+  let Inst{21-20} = 0b01;
+}
+
+def MVE_VSRIimm32 : MVE_VSxI_imm<"vsri", "32", 0b0, (ins shr_imm32:$imm)> {
+  let Inst{21} = 0b1;
+}
+
+def MVE_VSLIimm8 : MVE_VSxI_imm<"vsli", "8", 0b1, (ins imm0_7:$imm)> {
+  let Inst{21-19} = 0b001;
+}
+
+def MVE_VSLIimm16 : MVE_VSxI_imm<"vsli", "16", 0b1, (ins imm0_15:$imm)> {
+  let Inst{21-20} = 0b01;
+}
+
+def MVE_VSLIimm32 : MVE_VSxI_imm<"vsli", "32", 0b1,(ins imm0_31:$imm)> {
+  let Inst{21} = 0b1;
+}
+
+class MVE_VQSHL_imm<string suffix, dag imm>
+  : MVE_shift_with_imm<"vqshl", suffix, (outs MQPR:$Qd),
+                       !con((ins MQPR:$Qm), imm), "$Qd, $Qm, $imm",
+                       vpred_r, ""> {
+  bits<6> imm;
+
+  let Inst{25-24} = 0b11;
+  let Inst{21-16} = imm;
+  let Inst{10-8} = 0b111;
+}
+
+def MVE_VSLIimms8 : MVE_VQSHL_imm<"s8", (ins imm0_7:$imm)> {
+  let Inst{28} = 0b0;
+  let Inst{21-19} = 0b001;
+}
+
+def MVE_VSLIimmu8 : MVE_VQSHL_imm<"u8", (ins imm0_7:$imm)> {
+  let Inst{28} = 0b1;
+  let Inst{21-19} = 0b001;
+}
+
+def MVE_VSLIimms16 : MVE_VQSHL_imm<"s16", (ins imm0_15:$imm)> {
+  let Inst{28} = 0b0;
+  let Inst{21-20} = 0b01;
+}
+
+def MVE_VSLIimmu16 : MVE_VQSHL_imm<"u16", (ins imm0_15:$imm)> {
+  let Inst{28} = 0b1;
+  let Inst{21-20} = 0b01;
+}
+
+def MVE_VSLIimms32 : MVE_VQSHL_imm<"s32", (ins imm0_31:$imm)> {
+  let Inst{28} = 0b0;
+  let Inst{21} = 0b1;
+}
+
+def MVE_VSLIimmu32 : MVE_VQSHL_imm<"u32", (ins imm0_31:$imm)> {
+  let Inst{28} = 0b1;
+  let Inst{21} = 0b1;
+}
+
+class MVE_VQSHLU_imm<string suffix, dag imm>
+  : MVE_shift_with_imm<"vqshlu", suffix, (outs MQPR:$Qd),
+                       !con((ins MQPR:$Qm), imm), "$Qd, $Qm, $imm",
+                       vpred_r, ""> {
+  bits<6> imm;
+
+  let Inst{28} = 0b1;
+  let Inst{25-24} = 0b11;
+  let Inst{21-16} = imm;
+  let Inst{10-8} = 0b110;
+}
+
+def MVE_VQSHLU_imms8 : MVE_VQSHLU_imm<"s8", (ins imm0_7:$imm)> {
+  let Inst{21-19} = 0b001;
+}
+
+def MVE_VQSHLU_imms16 : MVE_VQSHLU_imm<"s16", (ins imm0_15:$imm)> {
+  let Inst{21-20} = 0b01;
+}
+
+def MVE_VQSHLU_imms32 : MVE_VQSHLU_imm<"s32", (ins imm0_31:$imm)> {
+  let Inst{21} = 0b1;
+}
+
+class MVE_VRSHR_imm<string suffix, dag imm>
+  : MVE_shift_with_imm<"vrshr", suffix, (outs MQPR:$Qd),
+                       !con((ins MQPR:$Qm), imm), "$Qd, $Qm, $imm",
+                       vpred_r, ""> {
+  bits<6> imm;
+
+  let Inst{25-24} = 0b11;
+  let Inst{21-16} = imm;
+  let Inst{10-8} = 0b010;
+}
+
+def MVE_VRSHR_imms8 : MVE_VRSHR_imm<"s8", (ins shr_imm8:$imm)> {
+  let Inst{28} = 0b0;
+  let Inst{21-19} = 0b001;
+}
+
+def MVE_VRSHR_immu8 : MVE_VRSHR_imm<"u8", (ins shr_imm8:$imm)> {
+  let Inst{28} = 0b1;
+  let Inst{21-19} = 0b001;
+}
+
+def MVE_VRSHR_imms16 : MVE_VRSHR_imm<"s16", (ins shr_imm16:$imm)> {
+  let Inst{28} = 0b0;
+  let Inst{21-20} = 0b01;
+}
+
+def MVE_VRSHR_immu16 : MVE_VRSHR_imm<"u16", (ins shr_imm16:$imm)> {
+  let Inst{28} = 0b1;
+  let Inst{21-20} = 0b01;
+}
+
+def MVE_VRSHR_imms32 : MVE_VRSHR_imm<"s32", (ins shr_imm32:$imm)> {
+  let Inst{28} = 0b0;
+  let Inst{21} = 0b1;
+}
+
+def MVE_VRSHR_immu32 : MVE_VRSHR_imm<"u32", (ins shr_imm32:$imm)> {
+  let Inst{28} = 0b1;
+  let Inst{21} = 0b1;
+}
+
+class MVE_VSHR_imm<string suffix, dag imm>
+  : MVE_shift_with_imm<"vshr", suffix, (outs MQPR:$Qd),
+                       !con((ins MQPR:$Qm), imm), "$Qd, $Qm, $imm",
+                       vpred_r, ""> {
+  bits<6> imm;
+
+  let Inst{25-24} = 0b11;
+  let Inst{21-16} = imm;
+  let Inst{10-8} = 0b000;
+}
+
+def MVE_VSHR_imms8 : MVE_VSHR_imm<"s8", (ins shr_imm8:$imm)> {
+  let Inst{28} = 0b0;
+  let Inst{21-19} = 0b001;
+}
+
+def MVE_VSHR_immu8 : MVE_VSHR_imm<"u8", (ins shr_imm8:$imm)> {
+  let Inst{28} = 0b1;
+  let Inst{21-19} = 0b001;
+}
+
+def MVE_VSHR_imms16 : MVE_VSHR_imm<"s16", (ins shr_imm16:$imm)> {
+  let Inst{28} = 0b0;
+  let Inst{21-20} = 0b01;
+}
+
+def MVE_VSHR_immu16 : MVE_VSHR_imm<"u16", (ins shr_imm16:$imm)> {
+  let Inst{28} = 0b1;
+  let Inst{21-20} = 0b01;
+}
+
+def MVE_VSHR_imms32 : MVE_VSHR_imm<"s32", (ins shr_imm32:$imm)> {
+  let Inst{28} = 0b0;
+  let Inst{21} = 0b1;
+}
+
+def MVE_VSHR_immu32 : MVE_VSHR_imm<"u32", (ins shr_imm32:$imm)> {
+  let Inst{28} = 0b1;
+  let Inst{21} = 0b1;
+}
+
+class MVE_VSHL_imm<string suffix, dag imm>
+  : MVE_shift_with_imm<"vshl", suffix, (outs MQPR:$Qd),
+                       !con((ins MQPR:$Qm), imm), "$Qd, $Qm, $imm",
+                       vpred_r, ""> {
+  bits<6> imm;
+
+  let Inst{28} = 0b0;
+  let Inst{25-24} = 0b11;
+  let Inst{21-16} = imm;
+  let Inst{10-8} = 0b101;
+}
+
+def MVE_VSHL_immi8 : MVE_VSHL_imm<"i8", (ins imm0_7:$imm)> {
+  let Inst{21-19} = 0b001;
+}
+
+def MVE_VSHL_immi16 : MVE_VSHL_imm<"i16", (ins imm0_15:$imm)> {
+  let Inst{21-20} = 0b01;
+}
+
+def MVE_VSHL_immi32 : MVE_VSHL_imm<"i32", (ins imm0_31:$imm)> {
+  let Inst{21} = 0b1;
+}
+// end of mve_shift instructions
+
 class MVE_VPT<string suffix, bits<2> size, dag iops, string asm, list<dag> pattern=[]>
   : MVE_MI<(outs ), iops, NoItinerary, !strconcat("vpt", "${Mk}", ".", suffix), asm, "", pattern> {
   bits<3> fc;
   bits<4> Mk;
   bits<3> Qn;
 
   let Inst{31-23} = 0b111111100;
   let Inst{22} = Mk{3};

llvm/trunk/lib/Target/ARM/AsmParser/ARMAsmParser.cpp

if ((Mnemonic == "movs" && isThumb()) ||
     return Mnemonic;
 
   // First, split out any predication code. Ignore mnemonics we know aren't
   // predicated but do have a carry-set and so weren't caught above.
   if (Mnemonic != "adcs" && Mnemonic != "bics" && Mnemonic != "movs" &&
       Mnemonic != "muls" && Mnemonic != "smlals" && Mnemonic != "smulls" &&
       Mnemonic != "umlals" && Mnemonic != "umulls" && Mnemonic != "lsls" &&
       Mnemonic != "sbcs" && Mnemonic != "rscs" &&
-      !(hasMVE() && Mnemonic == "vmine")) {
+      !(hasMVE() &&
+        (Mnemonic == "vmine" ||
+         Mnemonic == "vshle" || Mnemonic == "vshlt" || Mnemonic == "vshllt"))) {
     unsigned CC = ARMCondCodeFromString(Mnemonic.substr(Mnemonic.size()-2));
     if (CC != ~0U) {
       Mnemonic = Mnemonic.slice(0, Mnemonic.size() - 2);
       PredicationCode = CC;
     }
   }
 
   // Next, determine if we have a carry setting bit. We explicitly ignore all
unsigned IMod =
       .Case("id", ARM_PROC::ID)
       .Default(~0U);
     if (IMod != ~0U) {
       Mnemonic = Mnemonic.slice(0, Mnemonic.size()-2);
       ProcessorIMod = IMod;
     }
   }
 
-  if (isMnemonicVPTPredicable(Mnemonic, ExtraToken)) {
+  if (isMnemonicVPTPredicable(Mnemonic, ExtraToken) && Mnemonic != "vmovlt" &&
+      Mnemonic != "vshllt" && Mnemonic != "vrshrnt" && Mnemonic != "vshrnt" &&
+      Mnemonic != "vqrshrunt" && Mnemonic != "vqshrunt" &&
+      Mnemonic != "vqrshrnt" && Mnemonic != "vqshrnt") {
     unsigned CC = ARMVectorCondCodeFromString(Mnemonic.substr(Mnemonic.size()-1));
     if (CC != ~0U) {
       Mnemonic = Mnemonic.slice(0, Mnemonic.size()-1);
       VPTPredicationCode = CC;
     }
     return Mnemonic;
   }
 
bool ARMAsmParser::ParseInstruction(ParseInstructionInfo &Info, StringRef Name,
   if (CanAcceptPredicationCode) {
     SMLoc Loc = SMLoc::getFromPointer(NameLoc.getPointer() + Mnemonic.size() +
                                       CarrySetting);
     Operands.push_back(ARMOperand::CreateCondCode(
                        ARMCC::CondCodes(PredicationCode), Loc));
   }
 
   // Add the VPT predication code operand, if necessary.
-  if (CanAcceptVPTPredicationCode) {
+  // FIXME: We don't add them for the instructions filtered below as these can
+  // have custom operands which need special parsing.  This parsing requires
+  // the operand to be in the same place in the OperandVector as their
+  // definition in tblgen.  Since these instructions may also have the
+  // scalar predication operand we do not add the vector one and leave until
+  // now to fix it up.
+  if (CanAcceptVPTPredicationCode && Mnemonic != "vmov") {
     SMLoc Loc = SMLoc::getFromPointer(NameLoc.getPointer() + Mnemonic.size() +
                                       CarrySetting);
     Operands.push_back(ARMOperand::CreateVPTPred(
                          ARMVCC::VPTCodes(VPTPredicationCode), Loc));
   }
 
   // Add the processor imod operand, if necessary.
   if (ProcessorIMod) {
bool ARMAsmParser::ParseInstruction(ParseInstructionInfo &Info, StringRef Name,
   // appropriate predicate if needed.  This could be either the scalar
   // predication code or the vector predication code.
   if (PredicationCode == ARMCC::AL &&
       shouldOmitPredicateOperand(Mnemonic, Operands))
     Operands.erase(Operands.begin() + 1);
 
 
   if (hasMVE()) {
-    if (CanAcceptVPTPredicationCode) {
+    if (!shouldOmitVectorPredicateOperand(Mnemonic, Operands) &&
+        Mnemonic == "vmov" && PredicationCode == ARMCC::LT) {
+      // Very nasty hack to deal with the vector predicated variant of vmovlt
+      // the scalar predicated vmov with condition 'lt'.  We can not tell them
+      // apart until we have parsed their operands.
+      Operands.erase(Operands.begin() + 1);
+      Operands.erase(Operands.begin());
+      SMLoc MLoc = SMLoc::getFromPointer(NameLoc.getPointer());
+      SMLoc PLoc = SMLoc::getFromPointer(NameLoc.getPointer() +
+                                         Mnemonic.size() - 1 + CarrySetting);
+      Operands.insert(Operands.begin(),
+                      ARMOperand::CreateVPTPred(ARMVCC::None, PLoc));
+      Operands.insert(Operands.begin(),
+                      ARMOperand::CreateToken(StringRef("vmovlt"), MLoc));
+    } else if (CanAcceptVPTPredicationCode) {
       // For all other instructions, make sure only one of the two
       // predication operands is left behind, depending on whether we should
       // use the vector predication.
       if (shouldOmitVectorPredicateOperand(Mnemonic, Operands)) {
         if (CanAcceptPredicationCode)
           Operands.erase(Operands.begin() + 2);
         else
           Operands.erase(Operands.begin() + 1);
unsigned ARMAsmParser::validateTargetOperandClass(MCParsedAsmOperand &AsmOp,
   switch (Kind) {
   default: break;
   case MCK__35_0:
     if (Op.isImm())
       if (const MCConstantExpr *CE = dyn_cast<MCConstantExpr>(Op.getImm()))
         if (CE->getValue() == 0)
           return Match_Success;
     break;
+  case MCK__35_8:
+    if (Op.isImm())
+      if (const MCConstantExpr *CE = dyn_cast<MCConstantExpr>(Op.getImm()))
+        if (CE->getValue() == 8)
+          return Match_Success;
+    break;
+  case MCK__35_16:
+    if (Op.isImm())
+      if (const MCConstantExpr *CE = dyn_cast<MCConstantExpr>(Op.getImm()))
+        if (CE->getValue() == 16)
+          return Match_Success;
+    break;
   case MCK_ModImm:
     if (Op.isImm()) {
       const MCExpr *SOExpr = Op.getImm();
       int64_t Value;
       if (!SOExpr->evaluateAsAbsolute(Value))
         return Match_Success;
       assert((Value >= std::numeric_limits<int32_t>::min() &&
               Value <= std::numeric_limits<uint32_t>::max()) &&

llvm/trunk/lib/Target/ARM/MCTargetDesc/ARMMCCodeEmitter.cpp

public:
   uint32_t getT2AdrLabelOpValue(const MCInst &MI, unsigned OpIdx,
                               SmallVectorImpl<MCFixup> &Fixups,
                               const MCSubtargetInfo &STI) const;
 
   uint32_t getITMaskOpValue(const MCInst &MI, unsigned OpIdx,
                             SmallVectorImpl<MCFixup> &Fixups,
                             const MCSubtargetInfo &STI) const;
 
+  /// getMVEShiftImmOpValue - Return encoding info for the 'sz:imm5'
+  /// operand.
+  uint32_t getMVEShiftImmOpValue(const MCInst &MI, unsigned OpIdx,
+                                 SmallVectorImpl<MCFixup> &Fixups,
+                                 const MCSubtargetInfo &STI) const;
+
   /// getAddrModeImm12OpValue - Return encoding info for 'reg +/- imm12'
   /// operand.
   uint32_t getAddrModeImm12OpValue(const MCInst &MI, unsigned OpIdx,
                                    SmallVectorImpl<MCFixup> &Fixups,
                                    const MCSubtargetInfo &STI) const;
 
   /// getThumbAddrModeRegRegOpValue - Return encoding for 'reg + reg' operand.
   uint32_t getThumbAddrModeRegRegOpValue(const MCInst &MI, unsigned OpIdx,
getThumbAddrModeRegRegOpValue(const MCInst &MI, unsigned OpIdx,
   //   {2-0} = Rn
   const MCOperand &MO1 = MI.getOperand(OpIdx);
   const MCOperand &MO2 = MI.getOperand(OpIdx + 1);
   unsigned Rn = CTX.getRegisterInfo()->getEncodingValue(MO1.getReg());
   unsigned Rm = CTX.getRegisterInfo()->getEncodingValue(MO2.getReg());
   return (Rm << 3) | Rn;
 }
 
+/// getMVEShiftImmOpValue - Return encoding info for the 'sz:imm5'
+/// operand.
+uint32_t
+ARMMCCodeEmitter::getMVEShiftImmOpValue(const MCInst &MI, unsigned OpIdx,
+                                        SmallVectorImpl<MCFixup> &Fixups,
+                                        const MCSubtargetInfo &STI) const {
+  // {4-0} = szimm5
+  // The value we are trying to encode is an immediate between either the
+  // range of [1-7] or [1-15] depending on whether we are dealing with the
+  // u8/s8 or the u16/s16 variants respectively.
+  // This value is encoded as follows, if ShiftImm is the value within those
+  // ranges then the encoding szimm5 = ShiftImm + size, where size is either 8
+  // or 16.
+
+  unsigned Size, ShiftImm;
+  switch(MI.getOpcode()) {
+    case ARM::MVE_VSHLL_imms16bh:
+    case ARM::MVE_VSHLL_imms16th:
+    case ARM::MVE_VSHLL_immu16bh:
+    case ARM::MVE_VSHLL_immu16th:
+      Size = 16;
+      break;
+    case ARM::MVE_VSHLL_imms8bh:
+    case ARM::MVE_VSHLL_imms8th:
+    case ARM::MVE_VSHLL_immu8bh:
+    case ARM::MVE_VSHLL_immu8th:
+      Size = 8;
+      break;
+    default:
+      llvm_unreachable("Use of operand not supported by this instruction");
+  }
+  ShiftImm = MI.getOperand(OpIdx).getImm();
+  return Size + ShiftImm;
+}
+
 /// getAddrModeImm12OpValue - Return encoding info for 'reg +/- imm12' operand.
 uint32_t ARMMCCodeEmitter::
 getAddrModeImm12OpValue(const MCInst &MI, unsigned OpIdx,
                         SmallVectorImpl<MCFixup> &Fixups,
                         const MCSubtargetInfo &STI) const {
   // {17-13} = reg
   // {12}    = (U)nsigned (add == '1', sub == '0')
   // {11-0}  = imm12

llvm/trunk/test/MC/ARM/mve-shifts.s

+# RUN: not llvm-mc -triple=thumbv8.1m.main-none-eabi -mattr=+mve -show-encoding  < %s \
+# RUN:   | FileCheck --check-prefix=CHECK-NOFP %s
+# RUN: not llvm-mc -triple=thumbv8.1m.main-none-eabi -mattr=+mve.fp,+fp64 -show-encoding  < %s 2>%t \
+# RUN:   | FileCheck --check-prefix=CHECK %s
+# RUN:     FileCheck --check-prefix=ERROR < %t %s
+
+# CHECK: vshlc   q0, lr, #8  @ encoding: [0xa8,0xee,0xce,0x0f]
+# CHECK-NOFP: vshlc   q0, lr, #8  @ encoding: [0xa8,0xee,0xce,0x0f]
+vshlc   q0, lr, #8
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,32]
+vshlc   q0, lr, #33
+
+# CHECK: vmovlb.s8 q0, q6  @ encoding: [0xa8,0xee,0x4c,0x0f]
+# CHECK-NOFP: vmovlb.s8 q0, q6  @ encoding: [0xa8,0xee,0x4c,0x0f]
+vmovlb.s8 q0, q6
+
+# CHECK: vmovlt.s8 q0, q4  @ encoding: [0xa8,0xee,0x48,0x1f]
+# CHECK-NOFP: vmovlt.s8 q0, q4  @ encoding: [0xa8,0xee,0x48,0x1f]
+vmovlt.s8 q0, q4
+
+# CHECK: vpt.i8 eq, q0, q0
+# CHECK-NOFP: vpt.i8 eq, q0, q0
+# CHECK: vmovltt.s8 q0, q4  @ encoding: [0xa8,0xee,0x48,0x1f]
+# CHECK-NOFP: vmovltt.s8 q0, q4  @ encoding: [0xa8,0xee,0x48,0x1f]
+vpt.i8 eq, q0, q0
+vmovltt.s8 q0, q4
+
+# CHECK: vmovlb.u8 q0, q0  @ encoding: [0xa8,0xfe,0x40,0x0f]
+# CHECK-NOFP: vmovlb.u8 q0, q0  @ encoding: [0xa8,0xfe,0x40,0x0f]
+vmovlb.u8 q0, q0
+
+# CHECK: vmovlt.u8 q0, q2  @ encoding: [0xa8,0xfe,0x44,0x1f]
+# CHECK-NOFP: vmovlt.u8 q0, q2  @ encoding: [0xa8,0xfe,0x44,0x1f]
+vmovlt.u8 q0, q2
+
+# CHECK: vmovlb.u16 q1, q0  @ encoding: [0xb0,0xfe,0x40,0x2f]
+# CHECK-NOFP: vmovlb.u16 q1, q0  @ encoding: [0xb0,0xfe,0x40,0x2f]
+vmovlb.u16 q1, q0
+
+# CHECK: vmovlt.u16 q0, q2  @ encoding: [0xb0,0xfe,0x44,0x1f]
+# CHECK-NOFP: vmovlt.u16 q0, q2  @ encoding: [0xb0,0xfe,0x44,0x1f]
+vmovlt.u16 q0, q2
+
+# CHECK: vshllb.s8 q0, q2, #8  @ encoding: [0x31,0xee,0x05,0x0e]
+# CHECK-NOFP: vshllb.s8 q0, q2, #8  @ encoding: [0x31,0xee,0x05,0x0e]
+vshllb.s8 q0, q2, #8
+
+# CHECK: vshllt.s8 q1, q5, #8  @ encoding: [0x31,0xee,0x0b,0x3e]
+# CHECK-NOFP: vshllt.s8 q1, q5, #8  @ encoding: [0x31,0xee,0x0b,0x3e]
+vshllt.s8 q1, q5, #8
+
+# CHECK: vshllb.s8 q0, q0, #7  @ encoding: [0xaf,0xee,0x40,0x0f]
+# CHECK-NOFP: vshllb.s8 q0, q0, #7  @ encoding: [0xaf,0xee,0x40,0x0f]
+vshllb.s8 q0, q0, #7
+
+# CHECK: vshllb.u8 q1, q1, #8  @ encoding: [0x31,0xfe,0x03,0x2e]
+# CHECK-NOFP: vshllb.u8 q1, q1, #8  @ encoding: [0x31,0xfe,0x03,0x2e]
+vshllb.u8 q1, q1, #8
+
+# CHECK: vshllt.u8 q0, q0, #8  @ encoding: [0x31,0xfe,0x01,0x1e]
+# CHECK-NOFP: vshllt.u8 q0, q0, #8  @ encoding: [0x31,0xfe,0x01,0x1e]
+vshllt.u8 q0, q0, #8
+
+# CHECK: vshllb.u8 q0, q0, #3  @ encoding: [0xab,0xfe,0x40,0x0f]
+# CHECK-NOFP: vshllb.u8 q0, q0, #3  @ encoding: [0xab,0xfe,0x40,0x0f]
+vshllb.u8 q0, q0, #3
+
+# CHECK: vshllb.u16 q0, q5, #16  @ encoding: [0x35,0xfe,0x0b,0x0e]
+# CHECK-NOFP: vshllb.u16 q0, q5, #16  @ encoding: [0x35,0xfe,0x0b,0x0e]
+vshllb.u16 q0, q5, #16
+
+# CHECK: vshllt.u16 q0, q3, #16  @ encoding: [0x35,0xfe,0x07,0x1e]
+# CHECK-NOFP: vshllt.u16 q0, q3, #16  @ encoding: [0x35,0xfe,0x07,0x1e]
+vshllt.u16 q0, q3, #16
+
+# CHECK: vshllt.s16 q0, q0, #16  @ encoding: [0x35,0xee,0x01,0x1e]
+# CHECK-NOFP: vshllt.s16 q0, q0, #16  @ encoding: [0x35,0xee,0x01,0x1e]
+vshllt.s16 q0, q0, #16
+
+# CHECK: vshllt.s16 q0, q0, #14  @ encoding: [0xbe,0xee,0x40,0x1f]
+vshllt.s16 q0, q0, #14
+
+# CHECK: vshllt.s16 q0, q0, #11  @ encoding: [0xbb,0xee,0x40,0x1f]
+vshllt.s16 q0, q0, #11
+
+# CHECK: vshllb.u16 q0, q2, #4  @ encoding: [0xb4,0xfe,0x44,0x0f]
+# CHECK-NOFP: vshllb.u16 q0, q2, #4  @ encoding: [0xb4,0xfe,0x44,0x0f]
+vshllb.u16 q0, q2, #4
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,8]
+vshllb.s8 q0, q2, #9
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,8]
+vshllb.u8 q0, q2, #9
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,8]
+vshllb.u8 q0, q2, #0
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,16]
+vshllb.s16 q0, q2, #17
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,16]
+vshllb.u16 q0, q2, #17
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,16]
+vshllb.u16 q0, q2, #0
+
+# CHECK: vrshrnb.i16 q0, q3, #1  @ encoding: [0x8f,0xfe,0xc7,0x0f]
+# CHECK-NOFP: vrshrnb.i16 q0, q3, #1  @ encoding: [0x8f,0xfe,0xc7,0x0f]
+vrshrnb.i16 q0, q3, #1
+
+# CHECK: vrshrnt.i16 q0, q2, #5  @ encoding: [0x8b,0xfe,0xc5,0x1f]
+# CHECK-NOFP: vrshrnt.i16 q0, q2, #5  @ encoding: [0x8b,0xfe,0xc5,0x1f]
+vrshrnt.i16 q0, q2, #5
+
+# CHECK: vrshrnb.i32 q0, q4, #8  @ encoding: [0x98,0xfe,0xc9,0x0f]
+# CHECK-NOFP: vrshrnb.i32 q0, q4, #8  @ encoding: [0x98,0xfe,0xc9,0x0f]
+vrshrnb.i32 q0, q4, #8
+
+# CHECK: vrshrnt.i32 q0, q2, #7  @ encoding: [0x99,0xfe,0xc5,0x1f]
+# CHECK-NOFP: vrshrnt.i32 q0, q2, #7  @ encoding: [0x99,0xfe,0xc5,0x1f]
+vrshrnt.i32 q0, q2, #7
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,8]
+vrshrnb.i16 q0, q3, #9
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,16]
+vrshrnb.i32 q0, q3, #17
+
+# CHECK: vshrnb.i16 q1, q2, #1  @ encoding: [0x8f,0xee,0xc5,0x2f]
+# CHECK-NOFP: vshrnb.i16 q1, q2, #1  @ encoding: [0x8f,0xee,0xc5,0x2f]
+vshrnb.i16 q1, q2, #1
+
+# CHECK: vshrnt.i16 q0, q1, #1  @ encoding: [0x8f,0xee,0xc3,0x1f]
+# CHECK-NOFP: vshrnt.i16 q0, q1, #1  @ encoding: [0x8f,0xee,0xc3,0x1f]
+vshrnt.i16 q0, q1, #1
+
+# CHECK: vshrnb.i32 q0, q0, #12  @ encoding: [0x94,0xee,0xc1,0x0f]
+# CHECK-NOFP: vshrnb.i32 q0, q0, #12  @ encoding: [0x94,0xee,0xc1,0x0f]
+vshrnb.i32 q0, q0, #12
+
+# CHECK: vshrnt.i32 q0, q2, #4  @ encoding: [0x9c,0xee,0xc5,0x1f]
+# CHECK-NOFP: vshrnt.i32 q0, q2, #4  @ encoding: [0x9c,0xee,0xc5,0x1f]
+vshrnt.i32 q0, q2, #4
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,8]
+vshrnb.i16 q1, q2, #9
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,16]
+vshrnb.i32 q1, q2, #17
+
+# CHECK: vqrshrunb.s16 q0, q2, #8  @ encoding: [0x88,0xfe,0xc4,0x0f]
+# CHECK-NOFP: vqrshrunb.s16 q0, q2, #8  @ encoding: [0x88,0xfe,0xc4,0x0f]
+vqrshrunb.s16 q0, q2, #8
+
+# CHECK: vqrshrunt.s16 q0, q0, #6  @ encoding: [0x8a,0xfe,0xc0,0x1f]
+# CHECK-NOFP: vqrshrunt.s16 q0, q0, #6  @ encoding: [0x8a,0xfe,0xc0,0x1f]
+vqrshrunt.s16 q0, q0, #6
+
+# CHECK: vqrshrunt.s32 q0, q1, #8  @ encoding: [0x98,0xfe,0xc2,0x1f]
+# CHECK-NOFP: vqrshrunt.s32 q0, q1, #8  @ encoding: [0x98,0xfe,0xc2,0x1f]
+vqrshrunt.s32 q0, q1, #8
+
+# CHECK: vqrshrunb.s32 q0, q7, #13  @ encoding: [0x93,0xfe,0xce,0x0f]
+# CHECK-NOFP: vqrshrunb.s32 q0, q7, #13  @ encoding: [0x93,0xfe,0xce,0x0f]
+vqrshrunb.s32 q0, q7, #13
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,8]
+vqrshrunb.s16 q0, q2, #9
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,16]
+vqrshrunb.s32 q0, q2, #17
+
+# CHECK: vqshrunb.s16 q0, q7, #5  @ encoding: [0x8b,0xee,0xce,0x0f]
+# CHECK-NOFP: vqshrunb.s16 q0, q7, #5  @ encoding: [0x8b,0xee,0xce,0x0f]
+vqshrunb.s16 q0, q7, #5
+
+# CHECK: vqshrunt.s16 q0, q1, #7  @ encoding: [0x89,0xee,0xc2,0x1f]
+# CHECK-NOFP: vqshrunt.s16 q0, q1, #7  @ encoding: [0x89,0xee,0xc2,0x1f]
+vqshrunt.s16 q0, q1, #7
+
+# CHECK: vqshrunb.s32 q0, q6, #4  @ encoding: [0x9c,0xee,0xcc,0x0f]
+# CHECK-NOFP: vqshrunb.s32 q0, q6, #4  @ encoding: [0x9c,0xee,0xcc,0x0f]
+vqshrunb.s32 q0, q6, #4
+
+# CHECK: vqshrunt.s32 q0, q2, #10  @ encoding: [0x96,0xee,0xc4,0x1f]
+# CHECK-NOFP: vqshrunt.s32 q0, q2, #10  @ encoding: [0x96,0xee,0xc4,0x1f]
+vqshrunt.s32 q0, q2, #10
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,8]
+vqshrunt.s16 q0, q1, #9
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,16]
+vqshrunb.s32 q0, q6, #17
+
+# CHECK: vqrshrnb.s16 q0, q7, #8  @ encoding: [0x88,0xee,0x4f,0x0f]
+# CHECK-NOFP: vqrshrnb.s16 q0, q7, #8  @ encoding: [0x88,0xee,0x4f,0x0f]
+vqrshrnb.s16 q0, q7, #8
+
+# CHECK: vqrshrnt.u16 q1, q3, #4  @ encoding: [0x8c,0xfe,0x47,0x3f]
+# CHECK-NOFP: vqrshrnt.u16 q1, q3, #4  @ encoding: [0x8c,0xfe,0x47,0x3f]
+vqrshrnt.u16 q1, q3, #4
+
+# CHECK: vqrshrnb.u32 q0, q1, #7  @ encoding: [0x99,0xfe,0x43,0x0f]
+# CHECK-NOFP: vqrshrnb.u32 q0, q1, #7  @ encoding: [0x99,0xfe,0x43,0x0f]
+vqrshrnb.u32 q0, q1, #7
+
+# CHECK: vqrshrnt.s32 q0, q1, #11  @ encoding: [0x95,0xee,0x43,0x1f]
+# CHECK-NOFP: vqrshrnt.s32 q0, q1, #11  @ encoding: [0x95,0xee,0x43,0x1f]
+vqrshrnt.s32 q0, q1, #11
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,8]
+vqrshrnb.s16 q0, q7, #9
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,16]
+vqrshrnb.s32 q0, q7, #17
+
+# CHECK: vqshrnb.s16 q0, q6, #5  @ encoding: [0x8b,0xee,0x4c,0x0f]
+# CHECK-NOFP: vqshrnb.s16 q0, q6, #5  @ encoding: [0x8b,0xee,0x4c,0x0f]
+vqshrnb.s16 q0, q6, #5
+
+# CHECK: vqshrnt.s16 q0, q1, #4  @ encoding: [0x8c,0xee,0x42,0x1f]
+# CHECK-NOFP: vqshrnt.s16 q0, q1, #4  @ encoding: [0x8c,0xee,0x42,0x1f]
+vqshrnt.s16 q0, q1, #4
+
+# CHECK: vqshrnb.u16 q0, q3, #7  @ encoding: [0x89,0xfe,0x46,0x0f]
+# CHECK-NOFP: vqshrnb.u16 q0, q3, #7  @ encoding: [0x89,0xfe,0x46,0x0f]
+vqshrnb.u16 q0, q3, #7
+
+# CHECK: vqshrnt.u16 q0, q2, #8  @ encoding: [0x88,0xfe,0x44,0x1f]
+# CHECK-NOFP: vqshrnt.u16 q0, q2, #8  @ encoding: [0x88,0xfe,0x44,0x1f]
+vqshrnt.u16 q0, q2, #8
+
+# CHECK: vqshrnt.s32 q1, q4, #3  @ encoding: [0x9d,0xee,0x48,0x3f]
+# CHECK-NOFP: vqshrnt.s32 q1, q4, #3  @ encoding: [0x9d,0xee,0x48,0x3f]
+vqshrnt.s32 q1, q4, #3
+
+# CHECK: vqshrnb.u32 q0, q2, #14  @ encoding: [0x92,0xfe,0x44,0x0f]
+# CHECK-NOFP: vqshrnb.u32 q0, q2, #14  @ encoding: [0x92,0xfe,0x44,0x0f]
+vqshrnb.u32 q0, q2, #14
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,8]
+vqshrnb.s16 q0, q6, #9
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,16]
+vqshrnb.u32 q0, q6, #17
+
+# CHECK: vshl.s8 q6, q6, q6  @ encoding: [0x0c,0xef,0x4c,0xc4]
+# CHECK-NOFP: vshl.s8 q6, q6, q6  @ encoding: [0x0c,0xef,0x4c,0xc4]
+vshl.s8 q6, q6, q6
+
+# CHECK: vshl.s16 q0, q4, q2  @ encoding: [0x14,0xef,0x48,0x04]
+# CHECK-NOFP: vshl.s16 q0, q4, q2  @ encoding: [0x14,0xef,0x48,0x04]
+vshl.s16 q0, q4, q2
+
+# CHECK: vshl.s32 q1, q1, q5  @ encoding: [0x2a,0xef,0x42,0x24]
+# CHECK-NOFP: vshl.s32 q1, q1, q5  @ encoding: [0x2a,0xef,0x42,0x24]
+vshl.s32 q1, q1, q5
+
+# CHECK: vshl.u8 q1, q7, q2  @ encoding: [0x04,0xff,0x4e,0x24]
+# CHECK-NOFP: vshl.u8 q1, q7, q2  @ encoding: [0x04,0xff,0x4e,0x24]
+vshl.u8 q1, q7, q2
+
+# CHECK: vshl.u16 q0, q4, q0  @ encoding: [0x10,0xff,0x48,0x04]
+# CHECK-NOFP: vshl.u16 q0, q4, q0  @ encoding: [0x10,0xff,0x48,0x04]
+vshl.u16 q0, q4, q0
+
+# CHECK: vshl.u32 q2, q2, q4  @ encoding: [0x28,0xff,0x44,0x44]
+# CHECK-NOFP: vshl.u32 q2, q2, q4  @ encoding: [0x28,0xff,0x44,0x44]
+vshl.u32 q2, q2, q4
+
+# CHECK: vqshl.s8 q0, q1, q6  @ encoding: [0x0c,0xef,0x52,0x04]
+# CHECK-NOFP: vqshl.s8 q0, q1, q6  @ encoding: [0x0c,0xef,0x52,0x04]
+vqshl.s8 q0, q1, q6
+
+# CHECK: vqshl.s16 q4, q3, q7  @ encoding: [0x1e,0xef,0x56,0x84]
+# CHECK-NOFP: vqshl.s16 q4, q3, q7  @ encoding: [0x1e,0xef,0x56,0x84]
+vqshl.s16 q4, q3, q7
+
+# CHECK: vqshl.s32 q0, q5, q5  @ encoding: [0x2a,0xef,0x5a,0x04]
+# CHECK-NOFP: vqshl.s32 q0, q5, q5  @ encoding: [0x2a,0xef,0x5a,0x04]
+vqshl.s32 q0, q5, q5
+
+# CHECK: vqshl.u8 q0, q0, q6  @ encoding: [0x0c,0xff,0x50,0x04]
+# CHECK-NOFP: vqshl.u8 q0, q0, q6  @ encoding: [0x0c,0xff,0x50,0x04]
+vqshl.u8 q0, q0, q6
+
+# CHECK: vqshl.u16 q0, q5, q4  @ encoding: [0x18,0xff,0x5a,0x04]
+# CHECK-NOFP: vqshl.u16 q0, q5, q4  @ encoding: [0x18,0xff,0x5a,0x04]
+vqshl.u16 q0, q5, q4
+
+# CHECK: vqshl.u32 q1, q0, q4  @ encoding: [0x28,0xff,0x50,0x24]
+# CHECK-NOFP: vqshl.u32 q1, q0, q4  @ encoding: [0x28,0xff,0x50,0x24]
+vqshl.u32 q1, q0, q4
+
+# CHECK: vqrshl.s8 q1, q6, q1  @ encoding: [0x02,0xef,0x5c,0x25]
+# CHECK-NOFP: vqrshl.s8 q1, q6, q1  @ encoding: [0x02,0xef,0x5c,0x25]
+vqrshl.s8 q1, q6, q1
+
+# CHECK: vqrshl.s16 q2, q4, q6  @ encoding: [0x1c,0xef,0x58,0x45]
+# CHECK-NOFP: vqrshl.s16 q2, q4, q6  @ encoding: [0x1c,0xef,0x58,0x45]
+vqrshl.s16 q2, q4, q6
+
+# CHECK: vqrshl.s32 q0, q0, q5  @ encoding: [0x2a,0xef,0x50,0x05]
+# CHECK-NOFP: vqrshl.s32 q0, q0, q5  @ encoding: [0x2a,0xef,0x50,0x05]
+vqrshl.s32 q0, q0, q5
+
+# CHECK: vqrshl.u8 q0, q2, q1  @ encoding: [0x02,0xff,0x54,0x05]
+# CHECK-NOFP: vqrshl.u8 q0, q2, q1  @ encoding: [0x02,0xff,0x54,0x05]
+vqrshl.u8 q0, q2, q1
+
+# CHECK: vqrshl.u16 q1, q6, q0  @ encoding: [0x10,0xff,0x5c,0x25]
+# CHECK-NOFP: vqrshl.u16 q1, q6, q0  @ encoding: [0x10,0xff,0x5c,0x25]
+vqrshl.u16 q1, q6, q0
+
+# CHECK: vqrshl.u32 q0, q0, q0  @ encoding: [0x20,0xff,0x50,0x05]
+# CHECK-NOFP: vqrshl.u32 q0, q0, q0  @ encoding: [0x20,0xff,0x50,0x05]
+vqrshl.u32 q0, q0, q0
+
+# CHECK: vrshl.s8 q0, q6, q4  @ encoding: [0x08,0xef,0x4c,0x05]
+# CHECK-NOFP: vrshl.s8 q0, q6, q4  @ encoding: [0x08,0xef,0x4c,0x05]
+vrshl.s8 q0, q6, q4
+
+# CHECK: vrshl.s16 q1, q4, q7  @ encoding: [0x1e,0xef,0x48,0x25]
+# CHECK-NOFP: vrshl.s16 q1, q4, q7  @ encoding: [0x1e,0xef,0x48,0x25]
+vrshl.s16 q1, q4, q7
+
+# CHECK: vrshl.s32 q1, q4, q4  @ encoding: [0x28,0xef,0x48,0x25]
+# CHECK-NOFP: vrshl.s32 q1, q4, q4  @ encoding: [0x28,0xef,0x48,0x25]
+vrshl.s32 q1, q4, q4
+
+# CHECK: vrshl.u8 q0, q3, q5  @ encoding: [0x0a,0xff,0x46,0x05]
+# CHECK-NOFP: vrshl.u8 q0, q3, q5  @ encoding: [0x0a,0xff,0x46,0x05]
+vrshl.u8 q0, q3, q5
+
+# CHECK: vrshl.u16 q5, q6, q5  @ encoding: [0x1a,0xff,0x4c,0xa5]
+# CHECK-NOFP: vrshl.u16 q5, q6, q5  @ encoding: [0x1a,0xff,0x4c,0xa5]
+vrshl.u16 q5, q6, q5
+
+# CHECK: vrshl.u32 q1, q7, q3  @ encoding: [0x26,0xff,0x4e,0x25]
+# CHECK-NOFP: vrshl.u32 q1, q7, q3  @ encoding: [0x26,0xff,0x4e,0x25]
+vrshl.u32 q1, q7, q3
+
+# CHECK: vsri.8 q0, q2, #3  @ encoding: [0x8d,0xff,0x54,0x04]
+# CHECK-NOFP: vsri.8 q0, q2, #3  @ encoding: [0x8d,0xff,0x54,0x04]
+vsri.8 q0, q2, #3
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,8]
+vsri.8 q0, q2, #9
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,8]
+vsri.8 q0, q2, #0
+
+# CHECK: vsri.16 q0, q2, #5  @ encoding: [0x9b,0xff,0x54,0x04]
+# CHECK-NOFP: vsri.16 q0, q2, #5  @ encoding: [0x9b,0xff,0x54,0x04]
+vsri.16 q0, q2, #5
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,16]
+vsri.16 q0, q2, #17
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,16]
+vsri.16 q0, q2, #0
+
+# CHECK: vsri.32 q0, q1, #15  @ encoding: [0xb1,0xff,0x52,0x04]
+# CHECK-NOFP: vsri.32 q0, q1, #15  @ encoding: [0xb1,0xff,0x52,0x04]
+vsri.32 q0, q1, #15
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,32]
+vsri.32 q0, q2, #33
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,32]
+vsri.32 q0, q2, #0
+
+# CHECK: vsli.8 q0, q3, #3  @ encoding: [0x8b,0xff,0x56,0x05]
+# CHECK-NOFP: vsli.8 q0, q3, #3  @ encoding: [0x8b,0xff,0x56,0x05]
+vsli.8 q0, q3, #3
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [0,7]
+vsli.8 q0, q3, #8
+
+# CHECK: vsli.16 q0, q1, #12  @ encoding: [0x9c,0xff,0x52,0x05]
+# CHECK-NOFP: vsli.16 q0, q1, #12  @ encoding: [0x9c,0xff,0x52,0x05]
+vsli.16 q0, q1, #12
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [0,15]
+vsli.16 q0, q3, #16
+
+# CHECK: vsli.32 q0, q1, #8  @ encoding: [0xa8,0xff,0x52,0x05]
+# CHECK-NOFP: vsli.32 q0, q1, #8  @ encoding: [0xa8,0xff,0x52,0x05]
+vsli.32 q0, q1, #8
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [0,31]
+vsli.32 q0, q1, #32
+
+# CHECK: vqshl.s8 q0, q4, #6  @ encoding: [0x8e,0xef,0x58,0x07]
+# CHECK-NOFP: vqshl.s8 q0, q4, #6  @ encoding: [0x8e,0xef,0x58,0x07]
+vqshl.s8 q0, q4, #6
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [0,7]
+vqshl.s8 q0, q4, #8
+
+# CHECK: vqshl.u8 q0, q6, #6  @ encoding: [0x8e,0xff,0x5c,0x07]
+# CHECK-NOFP: vqshl.u8 q0, q6, #6  @ encoding: [0x8e,0xff,0x5c,0x07]
+vqshl.u8 q0, q6, #6
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [0,7]
+vqshl.u8 q0, q4, #8
+
+# CHECK: vqshl.s16 q1, q2, #5  @ encoding: [0x95,0xef,0x54,0x27]
+# CHECK-NOFP: vqshl.s16 q1, q2, #5  @ encoding: [0x95,0xef,0x54,0x27]
+vqshl.s16 q1, q2, #5
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [0,15]
+vqshl.s16 q1, q2, #16
+
+# CHECK: vqshl.u16 q0, q5, #3  @ encoding: [0x93,0xff,0x5a,0x07]
+# CHECK-NOFP: vqshl.u16 q0, q5, #3  @ encoding: [0x93,0xff,0x5a,0x07]
+vqshl.u16 q0, q5, #3
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [0,15]
+vqshl.u16 q1, q2, #16
+
+# CHECK: vqshl.s32 q1, q3, #29  @ encoding: [0xbd,0xef,0x56,0x27]
+# CHECK-NOFP: vqshl.s32 q1, q3, #29  @ encoding: [0xbd,0xef,0x56,0x27]
+vqshl.s32 q1, q3, #29
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [0,31]
+vqshl.s32 q1, q3, #32
+
+# CHECK: vqshl.u32 q0, q2, #19  @ encoding: [0xb3,0xff,0x54,0x07]
+# CHECK-NOFP: vqshl.u32 q0, q2, #19  @ encoding: [0xb3,0xff,0x54,0x07]
+vqshl.u32 q0, q2, #19
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [0,31]
+vqshl.u32 q0, q2, #32
+
+# CHECK: vqshlu.s8 q0, q1, #0  @ encoding: [0x88,0xff,0x52,0x06]
+# CHECK-NOFP: vqshlu.s8 q0, q1, #0  @ encoding: [0x88,0xff,0x52,0x06]
+vqshlu.s8 q0, q1, #0
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [0,7]
+vqshlu.s8 q0, q1, #8
+
+# CHECK: vqshlu.s16 q2, q1, #12  @ encoding: [0x9c,0xff,0x52,0x46]
+# CHECK-NOFP: vqshlu.s16 q2, q1, #12  @ encoding: [0x9c,0xff,0x52,0x46]
+vqshlu.s16 q2, q1, #12
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [0,15]
+vqshlu.s16 q0, q1, #16
+
+# CHECK: vqshlu.s32 q0, q4, #26  @ encoding: [0xba,0xff,0x58,0x06]
+# CHECK-NOFP: vqshlu.s32 q0, q4, #26  @ encoding: [0xba,0xff,0x58,0x06]
+vqshlu.s32 q0, q4, #26
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [0,31]
+vqshlu.s32 q0, q1, #32
+
+# CHECK: vrshr.s8 q1, q3, #7  @ encoding: [0x89,0xef,0x56,0x22]
+# CHECK-NOFP: vrshr.s8 q1, q3, #7  @ encoding: [0x89,0xef,0x56,0x22]
+vrshr.s8 q1, q3, #7
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,8]
+vrshr.s8 q1, q3, #9
+
+# CHECK: vrshr.u8 q1, q3, #2  @ encoding: [0x8e,0xff,0x56,0x22]
+# CHECK-NOFP: vrshr.u8 q1, q3, #2  @ encoding: [0x8e,0xff,0x56,0x22]
+vrshr.u8 q1, q3, #2
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,8]
+vrshr.u8 q1, q3, #9
+
+# CHECK: vrshr.s16 q0, q1, #10  @ encoding: [0x96,0xef,0x52,0x02]
+# CHECK-NOFP: vrshr.s16 q0, q1, #10  @ encoding: [0x96,0xef,0x52,0x02]
+vrshr.s16 q0, q1, #10
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,16]
+vrshr.s16 q0, q1, #17
+
+# CHECK: vrshr.u16 q0, q5, #12  @ encoding: [0x94,0xff,0x5a,0x02]
+# CHECK-NOFP: vrshr.u16 q0, q5, #12  @ encoding: [0x94,0xff,0x5a,0x02]
+vrshr.u16 q0, q5, #12
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,16]
+vrshr.u16 q0, q5, #20
+
+# CHECK: vrshr.s32 q0, q5, #23  @ encoding: [0xa9,0xef,0x5a,0x02]
+# CHECK-NOFP: vrshr.s32 q0, q5, #23  @ encoding: [0xa9,0xef,0x5a,0x02]
+vrshr.s32 q0, q5, #23
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,32]
+vrshr.s32 q0, q5, #33
+
+# CHECK: vrshr.u32 q0, q1, #30  @ encoding: [0xa2,0xff,0x52,0x02]
+# CHECK-NOFP: vrshr.u32 q0, q1, #30  @ encoding: [0xa2,0xff,0x52,0x02]
+vrshr.u32 q0, q1, #30
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,32]
+vrshr.u32 q0, q1, #55
+
+# CHECK: vshr.s8 q0, q7, #4  @ encoding: [0x8c,0xef,0x5e,0x00]
+# CHECK-NOFP: vshr.s8 q0, q7, #4  @ encoding: [0x8c,0xef,0x5e,0x00]
+vshr.s8 q0, q7, #4
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,8]
+vshr.s8 q0, q7, #9
+
+# CHECK: vshr.u8 q0, q2, #5  @ encoding: [0x8b,0xff,0x54,0x00]
+# CHECK-NOFP: vshr.u8 q0, q2, #5  @ encoding: [0x8b,0xff,0x54,0x00]
+vshr.u8 q0, q2, #5
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,8]
+vshr.u8 q0, q2, #9
+
+# CHECK: vshr.s16 q0, q3, #16  @ encoding: [0x90,0xef,0x56,0x00]
+# CHECK-NOFP: vshr.s16 q0, q3, #16  @ encoding: [0x90,0xef,0x56,0x00]
+vshr.s16 q0, q3, #16
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,16]
+vshr.s16 q0, q2, #17
+
+# CHECK: vshr.u16 q7, q6, #8  @ encoding: [0x98,0xff,0x5c,0xe0]
+# CHECK-NOFP: vshr.u16 q7, q6, #8  @ encoding: [0x98,0xff,0x5c,0xe0]
+vshr.u16 q7, q6, #8
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,16]
+vshr.u16 q7, q6, #20
+
+# CHECK: vshr.s32 q0, q6, #24  @ encoding: [0xa8,0xef,0x5c,0x00]
+# CHECK-NOFP: vshr.s32 q0, q6, #24  @ encoding: [0xa8,0xef,0x5c,0x00]
+vshr.s32 q0, q6, #24
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,32]
+vshr.s32 q0, q6, #33
+
+# CHECK: vshr.u32 q2, q5, #30  @ encoding: [0xa2,0xff,0x5a,0x40]
+# CHECK-NOFP: vshr.u32 q2, q5, #30  @ encoding: [0xa2,0xff,0x5a,0x40]
+vshr.u32 q2, q5, #30
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [1,32]
+vshr.u32 q2, q5, #33
+
+# CHECK: vshl.i8 q0, q6, #6  @ encoding: [0x8e,0xef,0x5c,0x05]
+# CHECK-NOFP: vshl.i8 q0, q6, #6  @ encoding: [0x8e,0xef,0x5c,0x05]
+vshl.i8 q0, q6, #6
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [0,7]
+vshl.i8 q0, q6, #8
+
+# CHECK: vshl.i16 q1, q0, #12  @ encoding: [0x9c,0xef,0x50,0x25]
+# CHECK-NOFP: vshl.i16 q1, q0, #12  @ encoding: [0x9c,0xef,0x50,0x25]
+vshl.i16 q1, q0, #12
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [0,15]
+vshl.i16 q1, q0, #16
+
+# CHECK: vshl.i32 q2, q2, #26  @ encoding: [0xba,0xef,0x54,0x45]
+# CHECK-NOFP: vshl.i32 q2, q2, #26  @ encoding: [0xba,0xef,0x54,0x45]
+vshl.i32 q2, q2, #26
+
+# ERROR: [[@LINE+1]]:{{[0-9]+}}: {{error|note}}: operand must be an immediate in the range [0,31]
+vshl.i32 q2, q2, #33
+
+vshllt.s8 q0, q1, #1
+# CHECK: vshllt.s8 q0, q1, #1 @ encoding: [0xa9,0xee,0x42,0x1f]
+# CHECK-NOFP: vshllt.s8 q0, q1, #1 @ encoding: [0xa9,0xee,0x42,0x1f]
+
+vpste
+vshlltt.s16 q0, q1, #4
+vshllbe.u16 q0, q1, #8
+# CHECK: vpste @ encoding: [0x71,0xfe,0x4d,0x8f]
+# CHECK-NOFP: vpste @ encoding: [0x71,0xfe,0x4d,0x8f]
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOFP: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK: vshllbe.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOFP: vshllbe.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]

llvm/trunk/test/MC/Disassembler/ARM/mve-shifts.txt

+# RUN: llvm-mc -disassemble -triple=thumbv8.1m.main-none-eabi -mattr=+mve.fp,+fp64 -show-encoding %s | FileCheck %s
+# RUN: not llvm-mc -disassemble -triple=thumbv8.1m.main-none-eabi -show-encoding %s &> %t
+# RUN: FileCheck --check-prefix=CHECK-NOMVE < %t %s
+
+# CHECK: vshlc   q0, lr, #8  @ encoding: [0xa8,0xee,0xce,0x0f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xa8,0xee,0xce,0x0f]
+
+# CHECK: vmovlb.s8 q0, q6  @ encoding: [0xa8,0xee,0x4c,0x0f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xa8,0xee,0x4c,0x0f]
+
+# CHECK: vmovlt.s8 q0, q4  @ encoding: [0xa8,0xee,0x48,0x1f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xa8,0xee,0x48,0x1f]
+
+# CHECK: vmovlb.u8 q0, q0  @ encoding: [0xa8,0xfe,0x40,0x0f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xa8,0xfe,0x40,0x0f]
+
+# CHECK: vmovlt.u8 q0, q2  @ encoding: [0xa8,0xfe,0x44,0x1f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xa8,0xfe,0x44,0x1f]
+
+# CHECK: vmovlb.u16 q1, q0  @ encoding: [0xb0,0xfe,0x40,0x2f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xb0,0xfe,0x40,0x2f]
+
+# CHECK: vmovlt.u16 q0, q2  @ encoding: [0xb0,0xfe,0x44,0x1f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xb0,0xfe,0x44,0x1f]
+
+# CHECK: vshllb.s8 q0, q2, #8  @ encoding: [0x31,0xee,0x05,0x0e]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x31,0xee,0x05,0x0e]
+
+# CHECK: vshllt.s8 q1, q5, #8  @ encoding: [0x31,0xee,0x0b,0x3e]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x31,0xee,0x0b,0x3e]
+
+# CHECK: vshllb.s8 q0, q0, #7  @ encoding: [0xaf,0xee,0x40,0x0f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xaf,0xee,0x40,0x0f]
+
+# CHECK: vshllb.u8 q1, q1, #8  @ encoding: [0x31,0xfe,0x03,0x2e]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x31,0xfe,0x03,0x2e]
+
+# CHECK: vshllt.u8 q0, q0, #8  @ encoding: [0x31,0xfe,0x01,0x1e]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x31,0xfe,0x01,0x1e]
+
+# CHECK: vshllb.u8 q0, q0, #3  @ encoding: [0xab,0xfe,0x40,0x0f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xab,0xfe,0x40,0x0f]
+
+# CHECK: vshllb.u16 q0, q5, #16  @ encoding: [0x35,0xfe,0x0b,0x0e]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x35,0xfe,0x0b,0x0e]
+
+# CHECK: vshllt.u16 q0, q3, #16  @ encoding: [0x35,0xfe,0x07,0x1e]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x35,0xfe,0x07,0x1e]
+
+# CHECK: vshllt.s16 q0, q0, #16  @ encoding: [0x35,0xee,0x01,0x1e]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x35,0xee,0x01,0x1e]
+
+# CHECK: vshllt.s16 q0, q0, #14  @ encoding: [0xbe,0xee,0x40,0x1f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xbe,0xee,0x40,0x1f]
+
+# CHECK: vshllt.s16 q0, q0, #11  @ encoding: [0xbb,0xee,0x40,0x1f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xbb,0xee,0x40,0x1f]
+
+# CHECK: vshllb.u16 q0, q2, #4  @ encoding: [0xb4,0xfe,0x44,0x0f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xb4,0xfe,0x44,0x0f]
+
+# CHECK: vrshrnb.i16 q0, q3, #1  @ encoding: [0x8f,0xfe,0xc7,0x0f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x8f,0xfe,0xc7,0x0f]
+
+# CHECK: vrshrnt.i16 q0, q2, #5  @ encoding: [0x8b,0xfe,0xc5,0x1f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x8b,0xfe,0xc5,0x1f]
+
+# CHECK: vrshrnb.i32 q0, q4, #8  @ encoding: [0x98,0xfe,0xc9,0x0f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x98,0xfe,0xc9,0x0f]
+
+# CHECK: vrshrnt.i32 q0, q2, #7  @ encoding: [0x99,0xfe,0xc5,0x1f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x99,0xfe,0xc5,0x1f]
+
+# CHECK: vshrnb.i16 q1, q2, #1  @ encoding: [0x8f,0xee,0xc5,0x2f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x8f,0xee,0xc5,0x2f]
+
+# CHECK: vshrnt.i16 q0, q1, #1  @ encoding: [0x8f,0xee,0xc3,0x1f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x8f,0xee,0xc3,0x1f]
+
+# CHECK: vshrnb.i32 q0, q0, #12  @ encoding: [0x94,0xee,0xc1,0x0f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x94,0xee,0xc1,0x0f]
+
+# CHECK: vshrnt.i32 q0, q2, #4  @ encoding: [0x9c,0xee,0xc5,0x1f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x9c,0xee,0xc5,0x1f]
+
+# CHECK: vqrshrunb.s16 q0, q2, #8  @ encoding: [0x88,0xfe,0xc4,0x0f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x88,0xfe,0xc4,0x0f]
+
+# CHECK: vqrshrunt.s16 q0, q0, #6  @ encoding: [0x8a,0xfe,0xc0,0x1f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x8a,0xfe,0xc0,0x1f]
+
+# CHECK: vqrshrunt.s32 q0, q1, #8  @ encoding: [0x98,0xfe,0xc2,0x1f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x98,0xfe,0xc2,0x1f]
+
+# CHECK: vqrshrunb.s32 q0, q7, #13  @ encoding: [0x93,0xfe,0xce,0x0f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x93,0xfe,0xce,0x0f]
+
+# CHECK: vqshrunb.s16 q0, q7, #5  @ encoding: [0x8b,0xee,0xce,0x0f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x8b,0xee,0xce,0x0f]
+
+# CHECK: vqshrunt.s16 q0, q1, #7  @ encoding: [0x89,0xee,0xc2,0x1f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x89,0xee,0xc2,0x1f]
+
+# CHECK: vqshrunb.s32 q0, q6, #4  @ encoding: [0x9c,0xee,0xcc,0x0f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x9c,0xee,0xcc,0x0f]
+
+# CHECK: vqshrunt.s32 q0, q2, #10  @ encoding: [0x96,0xee,0xc4,0x1f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x96,0xee,0xc4,0x1f]
+
+# CHECK: vqrshrnb.s16 q0, q7, #8  @ encoding: [0x88,0xee,0x4f,0x0f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x88,0xee,0x4f,0x0f]
+
+# CHECK: vqrshrnt.u16 q1, q3, #4  @ encoding: [0x8c,0xfe,0x47,0x3f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x8c,0xfe,0x47,0x3f]
+
+# CHECK: vqrshrnb.u32 q0, q1, #7  @ encoding: [0x99,0xfe,0x43,0x0f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x99,0xfe,0x43,0x0f]
+
+# CHECK: vqrshrnt.s32 q0, q1, #11  @ encoding: [0x95,0xee,0x43,0x1f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x95,0xee,0x43,0x1f]
+
+# CHECK: vqshrnb.s16 q0, q6, #5  @ encoding: [0x8b,0xee,0x4c,0x0f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x8b,0xee,0x4c,0x0f]
+
+# CHECK: vqshrnt.s16 q0, q1, #4  @ encoding: [0x8c,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x8c,0xee,0x42,0x1f]
+
+# CHECK: vqshrnb.u16 q0, q3, #7  @ encoding: [0x89,0xfe,0x46,0x0f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x89,0xfe,0x46,0x0f]
+
+# CHECK: vqshrnt.u16 q0, q2, #8  @ encoding: [0x88,0xfe,0x44,0x1f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x88,0xfe,0x44,0x1f]
+
+# CHECK: vqshrnt.s32 q1, q4, #3  @ encoding: [0x9d,0xee,0x48,0x3f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x9d,0xee,0x48,0x3f]
+
+# CHECK: vqshrnb.u32 q0, q2, #14  @ encoding: [0x92,0xfe,0x44,0x0f]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x92,0xfe,0x44,0x0f]
+
+# CHECK: vshl.s8 q6, q6, q6  @ encoding: [0x0c,0xef,0x4c,0xc4]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x0c,0xef,0x4c,0xc4]
+
+# CHECK: vshl.s16 q0, q4, q2  @ encoding: [0x14,0xef,0x48,0x04]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x14,0xef,0x48,0x04]
+
+# CHECK: vshl.s32 q1, q1, q5  @ encoding: [0x2a,0xef,0x42,0x24]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x2a,0xef,0x42,0x24]
+
+# CHECK: vshl.u8 q1, q7, q2  @ encoding: [0x04,0xff,0x4e,0x24]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x04,0xff,0x4e,0x24]
+
+# CHECK: vshl.u16 q0, q4, q0  @ encoding: [0x10,0xff,0x48,0x04]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x10,0xff,0x48,0x04]
+
+# CHECK: vshl.u32 q2, q2, q4  @ encoding: [0x28,0xff,0x44,0x44]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x28,0xff,0x44,0x44]
+
+# CHECK: vqshl.s8 q0, q1, q6  @ encoding: [0x0c,0xef,0x52,0x04]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x0c,0xef,0x52,0x04]
+
+# CHECK: vqshl.s16 q4, q3, q7  @ encoding: [0x1e,0xef,0x56,0x84]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x1e,0xef,0x56,0x84]
+
+# CHECK: vqshl.s32 q0, q5, q5  @ encoding: [0x2a,0xef,0x5a,0x04]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x2a,0xef,0x5a,0x04]
+
+# CHECK: vqshl.u8 q0, q0, q6  @ encoding: [0x0c,0xff,0x50,0x04]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x0c,0xff,0x50,0x04]
+
+# CHECK: vqshl.u16 q0, q5, q4  @ encoding: [0x18,0xff,0x5a,0x04]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x18,0xff,0x5a,0x04]
+
+# CHECK: vqshl.u32 q1, q0, q4  @ encoding: [0x28,0xff,0x50,0x24]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x28,0xff,0x50,0x24]
+
+# CHECK: vqrshl.s8 q1, q6, q1  @ encoding: [0x02,0xef,0x5c,0x25]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x02,0xef,0x5c,0x25]
+
+# CHECK: vqrshl.s16 q2, q4, q6  @ encoding: [0x1c,0xef,0x58,0x45]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x1c,0xef,0x58,0x45]
+
+# CHECK: vqrshl.s32 q0, q0, q5  @ encoding: [0x2a,0xef,0x50,0x05]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x2a,0xef,0x50,0x05]
+
+# CHECK: vqrshl.u8 q0, q2, q1  @ encoding: [0x02,0xff,0x54,0x05]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x02,0xff,0x54,0x05]
+
+# CHECK: vqrshl.u16 q1, q6, q0  @ encoding: [0x10,0xff,0x5c,0x25]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x10,0xff,0x5c,0x25]
+
+# CHECK: vqrshl.u32 q0, q0, q0  @ encoding: [0x20,0xff,0x50,0x05]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x20,0xff,0x50,0x05]
+
+# CHECK: vrshl.s8 q0, q6, q4  @ encoding: [0x08,0xef,0x4c,0x05]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x08,0xef,0x4c,0x05]
+
+# CHECK: vrshl.s16 q1, q4, q7  @ encoding: [0x1e,0xef,0x48,0x25]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x1e,0xef,0x48,0x25]
+
+# CHECK: vrshl.s32 q1, q4, q4  @ encoding: [0x28,0xef,0x48,0x25]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x28,0xef,0x48,0x25]
+
+# CHECK: vrshl.u8 q0, q3, q5  @ encoding: [0x0a,0xff,0x46,0x05]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x0a,0xff,0x46,0x05]
+
+# CHECK: vrshl.u16 q5, q6, q5  @ encoding: [0x1a,0xff,0x4c,0xa5]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x1a,0xff,0x4c,0xa5]
+
+# CHECK: vrshl.u32 q1, q7, q3  @ encoding: [0x26,0xff,0x4e,0x25]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x26,0xff,0x4e,0x25]
+
+# CHECK: vsri.8 q0, q2, #3  @ encoding: [0x8d,0xff,0x54,0x04]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x8d,0xff,0x54,0x04]
+
+# CHECK: vsri.16 q0, q2, #5  @ encoding: [0x9b,0xff,0x54,0x04]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x9b,0xff,0x54,0x04]
+
+# CHECK: vsri.32 q0, q1, #15  @ encoding: [0xb1,0xff,0x52,0x04]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xb1,0xff,0x52,0x04]
+
+# CHECK: vsli.8 q0, q3, #3  @ encoding: [0x8b,0xff,0x56,0x05]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x8b,0xff,0x56,0x05]
+
+# CHECK: vsli.16 q0, q1, #12  @ encoding: [0x9c,0xff,0x52,0x05]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x9c,0xff,0x52,0x05]
+
+# CHECK: vsli.32 q0, q1, #8  @ encoding: [0xa8,0xff,0x52,0x05]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xa8,0xff,0x52,0x05]
+
+# CHECK: vqshl.s8 q0, q4, #6  @ encoding: [0x8e,0xef,0x58,0x07]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x8e,0xef,0x58,0x07]
+
+# CHECK: vqshl.u8 q0, q6, #6  @ encoding: [0x8e,0xff,0x5c,0x07]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x8e,0xff,0x5c,0x07]
+
+# CHECK: vqshl.s16 q1, q2, #5  @ encoding: [0x95,0xef,0x54,0x27]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x95,0xef,0x54,0x27]
+
+# CHECK: vqshl.u16 q0, q5, #3  @ encoding: [0x93,0xff,0x5a,0x07]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x93,0xff,0x5a,0x07]
+
+# CHECK: vqshl.s32 q1, q3, #29  @ encoding: [0xbd,0xef,0x56,0x27]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xbd,0xef,0x56,0x27]
+
+# CHECK: vqshl.u32 q0, q2, #19  @ encoding: [0xb3,0xff,0x54,0x07]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xb3,0xff,0x54,0x07]
+
+# CHECK: vqshlu.s8 q0, q1, #0  @ encoding: [0x88,0xff,0x52,0x06]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x88,0xff,0x52,0x06]
+
+# CHECK: vqshlu.s16 q2, q1, #12  @ encoding: [0x9c,0xff,0x52,0x46]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x9c,0xff,0x52,0x46]
+
+# CHECK: vqshlu.s32 q0, q4, #26  @ encoding: [0xba,0xff,0x58,0x06]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xba,0xff,0x58,0x06]
+
+# CHECK: vrshr.s8 q1, q3, #7  @ encoding: [0x89,0xef,0x56,0x22]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x89,0xef,0x56,0x22]
+
+# CHECK: vrshr.u8 q1, q3, #2  @ encoding: [0x8e,0xff,0x56,0x22]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x8e,0xff,0x56,0x22]
+
+# CHECK: vrshr.s16 q0, q1, #10  @ encoding: [0x96,0xef,0x52,0x02]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x96,0xef,0x52,0x02]
+
+# CHECK: vrshr.u16 q0, q5, #12  @ encoding: [0x94,0xff,0x5a,0x02]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x94,0xff,0x5a,0x02]
+
+# CHECK: vrshr.s32 q0, q5, #23  @ encoding: [0xa9,0xef,0x5a,0x02]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xa9,0xef,0x5a,0x02]
+
+# CHECK: vrshr.u32 q0, q1, #30  @ encoding: [0xa2,0xff,0x52,0x02]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xa2,0xff,0x52,0x02]
+
+# CHECK: vshr.s8 q0, q7, #4  @ encoding: [0x8c,0xef,0x5e,0x00]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x8c,0xef,0x5e,0x00]
+
+# CHECK: vshr.u8 q0, q2, #5  @ encoding: [0x8b,0xff,0x54,0x00]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x8b,0xff,0x54,0x00]
+
+# CHECK: vshr.s16 q0, q3, #16  @ encoding: [0x90,0xef,0x56,0x00]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x90,0xef,0x56,0x00]
+
+# CHECK: vshr.u16 q7, q6, #8  @ encoding: [0x98,0xff,0x5c,0xe0]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x98,0xff,0x5c,0xe0]
+
+# CHECK: vshr.s32 q0, q6, #24  @ encoding: [0xa8,0xef,0x5c,0x00]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xa8,0xef,0x5c,0x00]
+
+# CHECK: vshr.u32 q2, q5, #30  @ encoding: [0xa2,0xff,0x5a,0x40]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xa2,0xff,0x5a,0x40]
+
+# CHECK: vshl.i8 q0, q6, #6  @ encoding: [0x8e,0xef,0x5c,0x05]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x8e,0xef,0x5c,0x05]
+
+# CHECK: vshl.i16 q1, q0, #12  @ encoding: [0x9c,0xef,0x50,0x25]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0x9c,0xef,0x50,0x25]
+
+# CHECK: vshl.i32 q2, q2, #26  @ encoding: [0xba,0xef,0x54,0x45]
+# CHECK-NOMVE: [[@LINE+1]]:2: warning: invalid instruction encoding
+[0xba,0xef,0x54,0x45]
+
+# CHECK: vpst @ encoding: [0x71,0xfe,0x4d,0x0f]
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+[0x71,0xfe,0x4d,0x0f]
+[0xb4,0xee,0x42,0x1f]
+
+# CHECK: vpstt @ encoding: [0x31,0xfe,0x4d,0x8f]
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK: vshllbt.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+[0x31,0xfe,0x4d,0x8f]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]
+
+# CHECK: vpste @ encoding: [0x71,0xfe,0x4d,0x8f]
+# CHECK-NOMVE: [[@LINE+5]]:2: warning: invalid instruction encoding
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+4]]:2: warning: invalid instruction encoding
+# CHECK: vshllbe.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOMVE: [[@LINE+3]]:2: warning: invalid instruction encoding
+[0x71,0xfe,0x4d,0x8f]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]
+
+# CHECK: vpsttt @ encoding: [0x31,0xfe,0x4d,0x4f]
+# CHECK-NOMVE: [[@LINE+7]]:2: warning: invalid instruction encoding
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+6]]:2: warning: invalid instruction encoding
+# CHECK: vshllbt.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOMVE: [[@LINE+5]]:2: warning: invalid instruction encoding
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+4]]:2: warning: invalid instruction encoding
+[0x31,0xfe,0x4d,0x4f]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]
+[0xb4,0xee,0x42,0x1f]
+
+# CHECK: vpstte @ encoding: [0x31,0xfe,0x4d,0xcf]
+# CHECK-NOMVE: [[@LINE+7]]:2: warning: invalid instruction encoding
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+6]]:2: warning: invalid instruction encoding
+# CHECK: vshllbt.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOMVE: [[@LINE+5]]:2: warning: invalid instruction encoding
+# CHECK: vshllte.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+4]]:2: warning: invalid instruction encoding
+[0x31,0xfe,0x4d,0xcf]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]
+[0xb4,0xee,0x42,0x1f]
+
+# CHECK: vpstet @ encoding: [0x71,0xfe,0x4d,0xcf]
+# CHECK-NOMVE: [[@LINE+7]]:2: warning: invalid instruction encoding
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+6]]:2: warning: invalid instruction encoding
+# CHECK: vshllbe.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOMVE: [[@LINE+5]]:2: warning: invalid instruction encoding
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+4]]:2: warning: invalid instruction encoding
+[0x71,0xfe,0x4d,0xcf]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]
+[0xb4,0xee,0x42,0x1f]
+
+# CHECK: vpstee @ encoding: [0x71,0xfe,0x4d,0x4f]
+# CHECK-NOMVE: [[@LINE+7]]:2: warning: invalid instruction encoding
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+6]]:2: warning: invalid instruction encoding
+# CHECK: vshllbe.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOMVE: [[@LINE+5]]:2: warning: invalid instruction encoding
+# CHECK: vshllte.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+4]]:2: warning: invalid instruction encoding
+[0x71,0xfe,0x4d,0x4f]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]
+[0xb4,0xee,0x42,0x1f]
+
+# CHECK: vpstttt @ encoding: [0x31,0xfe,0x4d,0x2f]
+# CHECK-NOMVE: [[@LINE+9]]:2: warning: invalid instruction encoding
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+8]]:2: warning: invalid instruction encoding
+# CHECK: vshllbt.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOMVE: [[@LINE+7]]:2: warning: invalid instruction encoding
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+6]]:2: warning: invalid instruction encoding
+# CHECK: vshllbt.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOMVE: [[@LINE+5]]:2: warning: invalid instruction encoding
+[0x31,0xfe,0x4d,0x2f]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]
+
+# CHECK: vpsttte @ encoding: [0x31,0xfe,0x4d,0x6f]
+# CHECK-NOMVE: [[@LINE+9]]:2: warning: invalid instruction encoding
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+8]]:2: warning: invalid instruction encoding
+# CHECK: vshllbt.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOMVE: [[@LINE+7]]:2: warning: invalid instruction encoding
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+6]]:2: warning: invalid instruction encoding
+# CHECK: vshllbe.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOMVE: [[@LINE+5]]:2: warning: invalid instruction encoding
+[0x31,0xfe,0x4d,0x6f]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]
+
+# CHECK: vpsttet @ encoding: [0x31,0xfe,0x4d,0xef]
+# CHECK-NOMVE: [[@LINE+9]]:2: warning: invalid instruction encoding
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+8]]:2: warning: invalid instruction encoding
+# CHECK: vshllbt.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOMVE: [[@LINE+7]]:2: warning: invalid instruction encoding
+# CHECK: vshllte.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+6]]:2: warning: invalid instruction encoding
+# CHECK: vshllbt.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOMVE: [[@LINE+5]]:2: warning: invalid instruction encoding
+[0x31,0xfe,0x4d,0xef]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]
+
+# CHECK: vpsttee @ encoding: [0x31,0xfe,0x4d,0xaf]
+# CHECK-NOMVE: [[@LINE+9]]:2: warning: invalid instruction encoding
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+8]]:2: warning: invalid instruction encoding
+# CHECK: vshllbt.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOMVE: [[@LINE+7]]:2: warning: invalid instruction encoding
+# CHECK: vshllte.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+6]]:2: warning: invalid instruction encoding
+# CHECK: vshllbe.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOMVE: [[@LINE+5]]:2: warning: invalid instruction encoding
+[0x31,0xfe,0x4d,0xaf]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]
+
+# CHECK: vpstett @ encoding: [0x71,0xfe,0x4d,0xaf]
+# CHECK-NOMVE: [[@LINE+9]]:2: warning: invalid instruction encoding
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+8]]:2: warning: invalid instruction encoding
+# CHECK: vshllbe.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOMVE: [[@LINE+7]]:2: warning: invalid instruction encoding
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+6]]:2: warning: invalid instruction encoding
+# CHECK: vshllbt.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOMVE: [[@LINE+5]]:2: warning: invalid instruction encoding
+[0x71,0xfe,0x4d,0xaf]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]
+
+# CHECK: vpstete @ encoding: [0x71,0xfe,0x4d,0xef]
+# CHECK-NOMVE: [[@LINE+9]]:2: warning: invalid instruction encoding
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+8]]:2: warning: invalid instruction encoding
+# CHECK: vshllbe.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOMVE: [[@LINE+7]]:2: warning: invalid instruction encoding
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+6]]:2: warning: invalid instruction encoding
+# CHECK: vshllbe.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOMVE: [[@LINE+5]]:2: warning: invalid instruction encoding
+[0x71,0xfe,0x4d,0xef]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]
+
+# CHECK: vpsteee @ encoding: [0x71,0xfe,0x4d,0x2f]
+# CHECK-NOMVE: [[@LINE+9]]:2: warning: invalid instruction encoding
+# CHECK: vshlltt.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+8]]:2: warning: invalid instruction encoding
+# CHECK: vshllbe.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOMVE: [[@LINE+7]]:2: warning: invalid instruction encoding
+# CHECK: vshllte.s16 q0, q1, #4 @ encoding: [0xb4,0xee,0x42,0x1f]
+# CHECK-NOMVE: [[@LINE+6]]:2: warning: invalid instruction encoding
+# CHECK: vshllbe.u16 q0, q1, #8 @ encoding: [0xb8,0xfe,0x42,0x0f]
+# CHECK-NOMVE: [[@LINE+5]]:2: warning: invalid instruction encoding
+[0x71,0xfe,0x4d,0x2f]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]
+[0xb4,0xee,0x42,0x1f]
+[0xb8,0xfe,0x42,0x0f]