M4:   "%mm4",
	M5:   "%mm5",
	M6:   "%mm6",
	M7:   "%mm7",
	X0:   "%xmm0",
	X1:   "%xmm1",
	X2:   "%xmm2",
	X3:   "%xmm3",
	X4:   "%xmm4",
	X5:   "%xmm5",
	X6:   "%xmm6",
	X7:   "%xmm7",
	X8:   "%xmm8",
	X9:   "%xmm9",
	X10:  "%xmm10",
	X11:  "%xmm11",
	X12:  "%xmm12",
	X13:  "%xmm13",
	X14:  "%xmm14",
	X15:  "%xmm15",
	CS:   "%cs",
	SS:   "%ss",
	DS:   "%ds",
	ES:   "%es",

	W13
	W14
	W15
	W16
	W17
	W18
	W19
	W20
	W21
	W22
	W23
	W24
	W25
	W26
	W27
	W28
	W29
	W30
	WZR

	X0
	X1
	X2
	X3
	X4
	X5
	X6
	X7
	X8
	X9
	X10
	X11
	X12
	X13
	X14
	X15
	X16
	X17
	X18
	X19
	X20
	X21
	X22
	X23
	X24
	X25
	X26
	X27
	X28
	X29
	X30
	XZR

	B0

					BadValue: []string{"bar", "foo"},
				},
			},
		},
		{
			Key: "x6",
			Op:  selection.Exists,
		},
		{
			Key: "x7",
			Op:  selection.DoesNotExist,
		},
		{
			Key:  "x8",
			Op:   selection.Exists,
			Vals: sets.NewString("foo"),
			WantErr: field.ErrorList{
				&field.Error{
					Type:     field.ErrorTypeInvalid,
					Field:    "values",
					BadValue: []string{"foo"},

(Lsh(64|32|16|8)x16 <t> x y) => (LOCGR {s390x.GreaterOrEqual} <t> (SL(D|W|W|W) <t> x y) (MOVDconst [0]) (CMPWUconst (MOVHZreg y) [64]))
(Lsh(64|32|16|8)x8  <t> x y) => (LOCGR {s390x.GreaterOrEqual} <t> (SL(D|W|W|W) <t> x y) (MOVDconst [0]) (CMPWUconst (MOVBZreg y) [64]))

(Rsh(64|32)Ux64 <t> x y) => (LOCGR {s390x.GreaterOrEqual} <t> (SR(D|W) <t> x y) (MOVDconst [0]) (CMPUconst y [64]))

//     Hardware

#include "textflag.h"

#define B0 X0
#define B1 X1
#define B2 X2
#define B3 X3
#define B4 X4
#define B5 X5
#define B6 X6
#define B7 X7

#define ACC0 X8
#define ACC1 X9
#define ACCM X10

#define T0 X11
#define T1 X12
#define T2 X13
#define POLY X14
#define BSWAP X15

DATA bswapMask<>+0x00(SB)/8, $0x08090a0b0c0d0e0f

	var x3 C.jthrowable = 0
	_ = x3
	var x4 C.jstring = 0
	_ = x4
	var x5 C.jarray = 0
	_ = x5
	var x6 C.jbooleanArray = 0
	_ = x6
	var x7 C.jbyteArray = 0
	_ = x7
	var x8 C.jcharArray = 0
	_ = x8
	var x9 C.jshortArray = 0
	_ = x9
	var x10 C.jintArray = 0
	_ = x10
	var x11 C.jlongArray = 0
	_ = x11
	var x12 C.jfloatArray = 0
	_ = x12
	var x13 C.jdoubleArray = 0

	"SP",
	"BP",
	"SI",
	"DI",
	"R8",
	"R9",
	"R10",
	"R11",
	"R12",
	"R13",
	"g", // a.k.a. R14
	"R15",
	"X0",
	"X1",
	"X2",
	"X3",
	"X4",
	"X5",
	"X6",
	"X7",
	"X8",
	"X9",
	"X10",
	"X11",
	"X12",
	"X13",
	"X14",
	"X15", // constant 0 in ABIInternal

	// If you add registers, update asyncPreempt in runtime

	// pseudo-registers
	"SB",
}

#define shiftC1Left BYTE $0x66; BYTE $0x0f; BYTE $0x3a; BYTE $0x0f; BYTE $0xff; BYTE $0x08 // PALIGNR $8, X7, X7
#define shiftC2Left BYTE $0x66; BYTE $0x45; BYTE $0x0f; BYTE $0x3a; BYTE $0x0f; BYTE $0xc0; BYTE $0x08 // PALIGNR $8, X8, X8
#define shiftC3Left BYTE $0x66; BYTE $0x45; BYTE $0x0f; BYTE $0x3a; BYTE $0x0f; BYTE $0xf6; BYTE $0x08 // PALIGNR $8, X14, X14

#define state0		X1	// round intermediates and outputs
#define state1		X2
#define m0		X3	// m0, m1,... m4 -- round message temps
#define m1		X4
#define m2		X5
#define m3		X6
#define m4		X7
#define shufMask	X8	// input data endian conversion control mask
#define abefSave	X9	// digest hash vector inter-block buffer abef
#define cdghSave	X10	// digest hash vector inter-block buffer cdgh

  // Build the shape of the refined type, if lhs is unranked it
  // will be directly the shape of the refined type, otherwise we merged by
  // taking the most specialized. This combines `10x?x?` and `?x?x8` into
  // `10x?x8`.
  if (!lhs_shape_type.hasRank()) {
    if (rhs_shape_type.hasRank()) {
      shape.append(rhs_shape_type.getShape().begin(),
                   rhs_shape_type.getShape().end());
      refined_shape = true;