8187573: [s390] z/Architecture Vector Facility Support
authorlucy
Tue, 26 Sep 2017 17:06:06 +0200
changeset 47531 6481d7b6c31e
parent 47530 468f0664cb39
child 47533 b77eae96ddef
child 47535 de6bea412643
child 47537 5390057a4489
8187573: [s390] z/Architecture Vector Facility Support Reviewed-by: mdoerr
src/hotspot/cpu/s390/assembler_s390.hpp
src/hotspot/cpu/s390/assembler_s390.inline.hpp
src/hotspot/cpu/s390/register_definitions_s390.cpp
src/hotspot/cpu/s390/register_s390.cpp
src/hotspot/cpu/s390/register_s390.hpp
--- a/src/hotspot/cpu/s390/assembler_s390.hpp	Tue Sep 26 06:53:35 2017 -0400
+++ b/src/hotspot/cpu/s390/assembler_s390.hpp	Tue Sep 26 17:06:06 2017 +0200
@@ -1,6 +1,6 @@
 /*
- * Copyright (c) 2016, Oracle and/or its affiliates. All rights reserved.
- * Copyright (c) 2016 SAP SE. All rights reserved.
+ * Copyright (c) 2016, 2017, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2016, 2017 SAP SE. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -250,7 +250,6 @@
   bool is_RSform()  { return has_base() && !has_index() && is_disp12(); }
   bool is_RSYform() { return has_base() && !has_index() && is_disp20(); }
   bool is_RXform()  { return has_base() &&  has_index() && is_disp12(); }
-  bool is_RXEform() { return has_base() &&  has_index() && is_disp12(); }
   bool is_RXYform() { return has_base() &&  has_index() && is_disp20(); }
 
   bool uses(Register r) { return _base == r || _index == r; };
@@ -1093,7 +1092,194 @@
 #define TRTT_ZOPC   (unsigned  int)(0xb9 << 24 | 0x90 << 16)
 
 
-// Miscellaneous Operations
+//---------------------------
+//--  Vector Instructions  --
+//---------------------------
+
+//---<  Vector Support Instructions  >---
+
+//---  Load (memory)  ---
+
+#define VLM_ZOPC    (unsigned long)(0xe7L << 40 | 0x36L << 0)   // load full vreg range (n * 128 bit)
+#define VL_ZOPC     (unsigned long)(0xe7L << 40 | 0x06L << 0)   // load full vreg (128 bit)
+#define VLEB_ZOPC   (unsigned long)(0xe7L << 40 | 0x00L << 0)   // load vreg element (8 bit)
+#define VLEH_ZOPC   (unsigned long)(0xe7L << 40 | 0x01L << 0)   // load vreg element (16 bit)
+#define VLEF_ZOPC   (unsigned long)(0xe7L << 40 | 0x03L << 0)   // load vreg element (32 bit)
+#define VLEG_ZOPC   (unsigned long)(0xe7L << 40 | 0x02L << 0)   // load vreg element (64 bit)
+
+#define VLREP_ZOPC  (unsigned long)(0xe7L << 40 | 0x05L << 0)   // load and replicate into all vector elements
+#define VLLEZ_ZOPC  (unsigned long)(0xe7L << 40 | 0x04L << 0)   // load logical element and zero.
+
+// vector register gather
+#define VGEF_ZOPC   (unsigned long)(0xe7L << 40 | 0x13L << 0)   // gather element (32 bit), V1(M3) = [D2(V2(M3),B2)]
+#define VGEG_ZOPC   (unsigned long)(0xe7L << 40 | 0x12L << 0)   // gather element (64 bit), V1(M3) = [D2(V2(M3),B2)]
+// vector register scatter
+#define VSCEF_ZOPC  (unsigned long)(0xe7L << 40 | 0x1bL << 0)   // vector scatter element FW
+#define VSCEG_ZOPC  (unsigned long)(0xe7L << 40 | 0x1aL << 0)   // vector scatter element DW
+
+#define VLBB_ZOPC   (unsigned long)(0xe7L << 40 | 0x07L << 0)   // load vreg to block boundary (load to alignment).
+#define VLL_ZOPC    (unsigned long)(0xe7L << 40 | 0x37L << 0)   // load vreg with length.
+
+//---  Load (register)  ---
+
+#define VLR_ZOPC    (unsigned long)(0xe7L << 40 | 0x56L << 0)   // copy full vreg (128 bit)
+#define VLGV_ZOPC   (unsigned long)(0xe7L << 40 | 0x21L << 0)   // copy vreg element -> GR
+#define VLVG_ZOPC   (unsigned long)(0xe7L << 40 | 0x22L << 0)   // copy GR -> vreg element
+#define VLVGP_ZOPC  (unsigned long)(0xe7L << 40 | 0x62L << 0)   // copy GR2, GR3 (disjoint pair) -> vreg
+
+// vector register pack: cut in half the size the source vector elements
+#define VPK_ZOPC    (unsigned long)(0xe7L << 40 | 0x94L << 0)   // just cut
+#define VPKS_ZOPC   (unsigned long)(0xe7L << 40 | 0x97L << 0)   // saturate as signed values
+#define VPKLS_ZOPC  (unsigned long)(0xe7L << 40 | 0x95L << 0)   // saturate as unsigned values
+
+// vector register unpack: double in size the source vector elements
+#define VUPH_ZOPC   (unsigned long)(0xe7L << 40 | 0xd7L << 0)   // signed, left half of the source vector elements
+#define VUPLH_ZOPC  (unsigned long)(0xe7L << 40 | 0xd5L << 0)   // unsigned, left half of the source vector elements
+#define VUPL_ZOPC   (unsigned long)(0xe7L << 40 | 0xd6L << 0)   // signed, right half of the source vector elements
+#define VUPLL_ZOPC  (unsigned long)(0xe7L << 40 | 0xd4L << 0)   // unsigned, right half of the source vector element
+
+// vector register merge
+#define VMRH_ZOPC   (unsigned long)(0xe7L << 40 | 0x61L << 0)   // register merge high (left half of source registers)
+#define VMRL_ZOPC   (unsigned long)(0xe7L << 40 | 0x60L << 0)   // register merge low (right half of source registers)
+
+// vector register permute
+#define VPERM_ZOPC  (unsigned long)(0xe7L << 40 | 0x8cL << 0)   // vector permute
+#define VPDI_ZOPC   (unsigned long)(0xe7L << 40 | 0x84L << 0)   // vector permute DW immediate
+
+// vector register replicate
+#define VREP_ZOPC   (unsigned long)(0xe7L << 40 | 0x4dL << 0)   // vector replicate
+#define VREPI_ZOPC  (unsigned long)(0xe7L << 40 | 0x45L << 0)   // vector replicate immediate
+#define VSEL_ZOPC   (unsigned long)(0xe7L << 40 | 0x8dL << 0)   // vector select
+
+#define VSEG_ZOPC   (unsigned long)(0xe7L << 40 | 0x5fL << 0)   // vector sign-extend to DW (rightmost element in each DW).
+
+//---  Load (immediate)  ---
+
+#define VLEIB_ZOPC  (unsigned long)(0xe7L << 40 | 0x40L << 0)   // load vreg element (16 bit imm to 8 bit)
+#define VLEIH_ZOPC  (unsigned long)(0xe7L << 40 | 0x41L << 0)   // load vreg element (16 bit imm to 16 bit)
+#define VLEIF_ZOPC  (unsigned long)(0xe7L << 40 | 0x43L << 0)   // load vreg element (16 bit imm to 32 bit)
+#define VLEIG_ZOPC  (unsigned long)(0xe7L << 40 | 0x42L << 0)   // load vreg element (16 bit imm to 64 bit)
+
+//---  Store  ---
+
+#define VSTM_ZOPC   (unsigned long)(0xe7L << 40 | 0x3eL << 0)   // store full vreg range (n * 128 bit)
+#define VST_ZOPC    (unsigned long)(0xe7L << 40 | 0x0eL << 0)   // store full vreg (128 bit)
+#define VSTEB_ZOPC  (unsigned long)(0xe7L << 40 | 0x08L << 0)   // store vreg element (8 bit)
+#define VSTEH_ZOPC  (unsigned long)(0xe7L << 40 | 0x09L << 0)   // store vreg element (16 bit)
+#define VSTEF_ZOPC  (unsigned long)(0xe7L << 40 | 0x0bL << 0)   // store vreg element (32 bit)
+#define VSTEG_ZOPC  (unsigned long)(0xe7L << 40 | 0x0aL << 0)   // store vreg element (64 bit)
+#define VSTL_ZOPC   (unsigned long)(0xe7L << 40 | 0x3fL << 0)   // store vreg with length.
+
+//---  Misc  ---
+
+#define VGM_ZOPC    (unsigned long)(0xe7L << 40 | 0x46L << 0)   // generate bit  mask, [start..end] = '1', else '0'
+#define VGBM_ZOPC   (unsigned long)(0xe7L << 40 | 0x44L << 0)   // generate byte mask, bits(imm16) -> bytes
+
+//---<  Vector Arithmetic Instructions  >---
+
+// Load
+#define VLC_ZOPC    (unsigned long)(0xe7L << 40 | 0xdeL << 0)   // V1 := -V2,   element size = 2**m
+#define VLP_ZOPC    (unsigned long)(0xe7L << 40 | 0xdfL << 0)   // V1 := |V2|,  element size = 2**m
+
+// ADD
+#define VA_ZOPC     (unsigned long)(0xe7L << 40 | 0xf3L << 0)   // V1 := V2 + V3, element size = 2**m
+#define VACC_ZOPC   (unsigned long)(0xe7L << 40 | 0xf1L << 0)   // V1 := carry(V2 + V3), element size = 2**m
+
+// SUB
+#define VS_ZOPC     (unsigned long)(0xe7L << 40 | 0xf7L << 0)   // V1 := V2 - V3, element size = 2**m
+#define VSCBI_ZOPC  (unsigned long)(0xe7L << 40 | 0xf5L << 0)   // V1 := borrow(V2 - V3), element size = 2**m
+
+// MUL
+#define VML_ZOPC    (unsigned long)(0xe7L << 40 | 0xa2L << 0)   // V1 := V2 * V3, element size = 2**m
+#define VMH_ZOPC    (unsigned long)(0xe7L << 40 | 0xa3L << 0)   // V1 := V2 * V3, element size = 2**m
+#define VMLH_ZOPC   (unsigned long)(0xe7L << 40 | 0xa1L << 0)   // V1 := V2 * V3, element size = 2**m, unsigned
+#define VME_ZOPC    (unsigned long)(0xe7L << 40 | 0xa6L << 0)   // V1 := V2 * V3, element size = 2**m
+#define VMLE_ZOPC   (unsigned long)(0xe7L << 40 | 0xa4L << 0)   // V1 := V2 * V3, element size = 2**m, unsigned
+#define VMO_ZOPC    (unsigned long)(0xe7L << 40 | 0xa7L << 0)   // V1 := V2 * V3, element size = 2**m
+#define VMLO_ZOPC   (unsigned long)(0xe7L << 40 | 0xa5L << 0)   // V1 := V2 * V3, element size = 2**m, unsigned
+
+// MUL & ADD
+#define VMAL_ZOPC   (unsigned long)(0xe7L << 40 | 0xaaL << 0)   // V1 := V2 * V3 + V4, element size = 2**m
+#define VMAH_ZOPC   (unsigned long)(0xe7L << 40 | 0xabL << 0)   // V1 := V2 * V3 + V4, element size = 2**m
+#define VMALH_ZOPC  (unsigned long)(0xe7L << 40 | 0xa9L << 0)   // V1 := V2 * V3 + V4, element size = 2**m, unsigned
+#define VMAE_ZOPC   (unsigned long)(0xe7L << 40 | 0xaeL << 0)   // V1 := V2 * V3 + V4, element size = 2**m
+#define VMALE_ZOPC  (unsigned long)(0xe7L << 40 | 0xacL << 0)   // V1 := V2 * V3 + V4, element size = 2**m, unsigned
+#define VMAO_ZOPC   (unsigned long)(0xe7L << 40 | 0xafL << 0)   // V1 := V2 * V3 + V4, element size = 2**m
+#define VMALO_ZOPC  (unsigned long)(0xe7L << 40 | 0xadL << 0)   // V1 := V2 * V3 + V4, element size = 2**m, unsigned
+
+// Vector SUM
+#define VSUM_ZOPC   (unsigned long)(0xe7L << 40 | 0x64L << 0)   // V1[j] := toFW(sum(V2[i]) + V3[j]), subelements: byte or HW
+#define VSUMG_ZOPC  (unsigned long)(0xe7L << 40 | 0x65L << 0)   // V1[j] := toDW(sum(V2[i]) + V3[j]), subelements: HW or FW
+#define VSUMQ_ZOPC  (unsigned long)(0xe7L << 40 | 0x67L << 0)   // V1[j] := toQW(sum(V2[i]) + V3[j]), subelements: FW or DW
+
+// Average
+#define VAVG_ZOPC   (unsigned long)(0xe7L << 40 | 0xf2L << 0)   // V1 := (V2+V3+1)/2, signed,   element size = 2**m
+#define VAVGL_ZOPC  (unsigned long)(0xe7L << 40 | 0xf0L << 0)   // V1 := (V2+V3+1)/2, unsigned, element size = 2**m
+
+// VECTOR Galois Field Multiply Sum
+#define VGFM_ZOPC   (unsigned long)(0xe7L << 40 | 0xb4L << 0)
+#define VGFMA_ZOPC  (unsigned long)(0xe7L << 40 | 0xbcL << 0)
+
+//---<  Vector Logical Instructions  >---
+
+// AND
+#define VN_ZOPC     (unsigned long)(0xe7L << 40 | 0x68L << 0)   // V1 := V2 & V3,  element size = 2**m
+#define VNC_ZOPC    (unsigned long)(0xe7L << 40 | 0x69L << 0)   // V1 := V2 & ~V3, element size = 2**m
+
+// XOR
+#define VX_ZOPC     (unsigned long)(0xe7L << 40 | 0x6dL << 0)   // V1 := V2 ^ V3,  element size = 2**m
+
+// NOR
+#define VNO_ZOPC    (unsigned long)(0xe7L << 40 | 0x6bL << 0)   // V1 := !(V2 | V3),  element size = 2**m
+
+// OR
+#define VO_ZOPC     (unsigned long)(0xe7L << 40 | 0x6aL << 0)   // V1 := V2 | V3,  element size = 2**m
+
+// Comparison (element-wise)
+#define VCEQ_ZOPC   (unsigned long)(0xe7L << 40 | 0xf8L << 0)   // V1 := (V2 == V3) ? 0xffff : 0x0000, element size = 2**m
+#define VCH_ZOPC    (unsigned long)(0xe7L << 40 | 0xfbL << 0)   // V1 := (V2  > V3) ? 0xffff : 0x0000, element size = 2**m, signed
+#define VCHL_ZOPC   (unsigned long)(0xe7L << 40 | 0xf9L << 0)   // V1 := (V2  > V3) ? 0xffff : 0x0000, element size = 2**m, unsigned
+
+// Max/Min (element-wise)
+#define VMX_ZOPC    (unsigned long)(0xe7L << 40 | 0xffL << 0)   // V1 := (V2 > V3) ? V2 : V3, element size = 2**m, signed
+#define VMXL_ZOPC   (unsigned long)(0xe7L << 40 | 0xfdL << 0)   // V1 := (V2 > V3) ? V2 : V3, element size = 2**m, unsigned
+#define VMN_ZOPC    (unsigned long)(0xe7L << 40 | 0xfeL << 0)   // V1 := (V2 < V3) ? V2 : V3, element size = 2**m, signed
+#define VMNL_ZOPC   (unsigned long)(0xe7L << 40 | 0xfcL << 0)   // V1 := (V2 < V3) ? V2 : V3, element size = 2**m, unsigned
+
+// Leading/Trailing Zeros, population count
+#define VCLZ_ZOPC   (unsigned long)(0xe7L << 40 | 0x53L << 0)   // V1 := leadingzeros(V2),  element size = 2**m
+#define VCTZ_ZOPC   (unsigned long)(0xe7L << 40 | 0x52L << 0)   // V1 := trailingzeros(V2), element size = 2**m
+#define VPOPCT_ZOPC (unsigned long)(0xe7L << 40 | 0x50L << 0)   // V1 := popcount(V2), bytewise!!
+
+// Rotate/Shift
+#define VERLLV_ZOPC (unsigned long)(0xe7L << 40 | 0x73L << 0)   // V1 := rotateleft(V2), rotate count in V3 element
+#define VERLL_ZOPC  (unsigned long)(0xe7L << 40 | 0x33L << 0)   // V1 := rotateleft(V3), rotate count from d2(b2).
+#define VERIM_ZOPC  (unsigned long)(0xe7L << 40 | 0x72L << 0)   // Rotate then insert under mask. Read Principles of Operation!!
+
+#define VESLV_ZOPC  (unsigned long)(0xe7L << 40 | 0x70L << 0)   // V1 := SLL(V2, V3), unsigned, element-wise
+#define VESL_ZOPC   (unsigned long)(0xe7L << 40 | 0x30L << 0)   // V1 := SLL(V3), unsigned, shift count from d2(b2).
+
+#define VESRAV_ZOPC (unsigned long)(0xe7L << 40 | 0x7AL << 0)   // V1 := SRA(V2, V3), signed, element-wise
+#define VESRA_ZOPC  (unsigned long)(0xe7L << 40 | 0x3AL << 0)   // V1 := SRA(V3), signed, shift count from d2(b2).
+#define VESRLV_ZOPC (unsigned long)(0xe7L << 40 | 0x78L << 0)   // V1 := SRL(V2, V3), unsigned, element-wise
+#define VESRL_ZOPC  (unsigned long)(0xe7L << 40 | 0x38L << 0)   // V1 := SRL(V3), unsigned, shift count from d2(b2).
+
+#define VSL_ZOPC    (unsigned long)(0xe7L << 40 | 0x74L << 0)   // V1 := SLL(V2), unsigned, bit-count
+#define VSLB_ZOPC   (unsigned long)(0xe7L << 40 | 0x75L << 0)   // V1 := SLL(V2), unsigned, byte-count
+#define VSLDB_ZOPC  (unsigned long)(0xe7L << 40 | 0x77L << 0)   // V1 := SLL((V2,V3)), unsigned, byte-count
+
+#define VSRA_ZOPC   (unsigned long)(0xe7L << 40 | 0x7eL << 0)   // V1 := SRA(V2), signed, bit-count
+#define VSRAB_ZOPC  (unsigned long)(0xe7L << 40 | 0x7fL << 0)   // V1 := SRA(V2), signed, byte-count
+#define VSRL_ZOPC   (unsigned long)(0xe7L << 40 | 0x7cL << 0)   // V1 := SRL(V2), unsigned, bit-count
+#define VSRLB_ZOPC  (unsigned long)(0xe7L << 40 | 0x7dL << 0)   // V1 := SRL(V2), unsigned, byte-count
+
+// Test under Mask
+#define VTM_ZOPC    (unsigned long)(0xe7L << 40 | 0xd8L << 0)   // Like TM, set CC according to state of selected bits.
+
+
+//--------------------------------
+//--  Miscellaneous Operations  --
+//--------------------------------
 
 // Execute
 #define EX_ZOPC     (unsigned  int)(68L << 24)
@@ -1280,6 +1466,21 @@
     to_minus_infinity = 7
   };
 
+  // Vector Register Element Type.
+  enum VRegElemType {
+    VRET_BYTE   = 0,
+    VRET_HW     = 1,
+    VRET_FW     = 2,
+    VRET_DW     = 3,
+    VRET_QW     = 4
+  };
+
+  // Vector Operation Condition Code Control.
+  enum VOpCCC {
+    VOP_CCIGN   = 0, // ignore, don't set CC
+    VOP_CCSET   = 1  // set the CC
+  };
+
   // Inverse condition code, i.e. determine "15 - cc" for a given condition code cc.
   static branch_condition inverse_condition(branch_condition cc);
   static branch_condition inverse_float_condition(branch_condition cc);
@@ -1376,6 +1577,60 @@
     return r;
   }
 
+  static int64_t rsmask_48( Address a) { assert(a.is_RSform(),  "bad address format"); return rsmask_48( a.disp12(), a.base()); }
+  static int64_t rxmask_48( Address a) {      if (a.is_RXform())  { return rxmask_48( a.disp12(), a.index(), a.base()); }
+                                         else if (a.is_RSform())  { return rsmask_48( a.disp12(),            a.base()); }
+                                         else                     { guarantee(false, "bad address format");  return 0;  }
+                                       }
+  static int64_t rsymask_48(Address a) { assert(a.is_RSYform(), "bad address format"); return rsymask_48(a.disp20(), a.base()); }
+  static int64_t rxymask_48(Address a) {      if (a.is_RXYform()) { return rxymask_48( a.disp20(), a.index(), a.base()); }
+                                         else if (a.is_RSYform()) { return rsymask_48( a.disp20(),            a.base()); }
+                                         else                     { guarantee(false, "bad address format");  return 0;   }
+                                       }
+
+  static int64_t rsmask_48( int64_t d2, Register b2)              { return uimm12(d2, 20, 48)                   | regz(b2, 16, 48); }
+  static int64_t rxmask_48( int64_t d2, Register x2, Register b2) { return uimm12(d2, 20, 48) | reg(x2, 12, 48) | regz(b2, 16, 48); }
+  static int64_t rsymask_48(int64_t d2, Register b2)              { return simm20(d2)                           | regz(b2, 16, 48); }
+  static int64_t rxymask_48(int64_t d2, Register x2, Register b2) { return simm20(d2)         | reg(x2, 12, 48) | regz(b2, 16, 48); }
+
+  // Address calculated from d12(vx,b) - vx is vector index register.
+  static int64_t rvmask_48( int64_t d2, VectorRegister x2, Register b2) { return uimm12(d2, 20, 48) | vreg(x2, 12) | regz(b2, 16, 48); }
+
+  static int64_t vreg_mask(VectorRegister v, int pos) {
+    return vreg(v, pos) | v->RXB_mask(pos);
+  }
+
+  // Vector Element Size Control. 4-bit field which indicates the size of the vector elements.
+  static int64_t vesc_mask(int64_t size, int min_size, int max_size, int pos) {
+    // min_size - minimum element size. Not all instructions support element sizes beginning with "byte".
+    // max_size - maximum element size. Not all instructions support element sizes up to "QW".
+    assert((min_size <= size) && (size <= max_size), "element size control out of range");
+    return uimm4(size, pos, 48);
+  }
+
+  // Vector Element IndeX. 4-bit field which indexes the target vector element.
+  static int64_t veix_mask(int64_t ix, int el_size, int pos) {
+    // el_size - size of the vector element. This is a VRegElemType enum value.
+    // ix      - vector element index.
+    int max_ix = -1;
+    switch (el_size) {
+      case VRET_BYTE: max_ix = 15; break;
+      case VRET_HW:   max_ix =  7; break;
+      case VRET_FW:   max_ix =  3; break;
+      case VRET_DW:   max_ix =  1; break;
+      case VRET_QW:   max_ix =  0; break;
+      default:        guarantee(false, "bad vector element size %d", el_size); break;
+    }
+    assert((0 <= ix) && (ix <= max_ix), "element size out of range (0 <= %ld <= %d)", ix, max_ix);
+    return uimm4(ix, pos, 48);
+  }
+
+  // Vector Operation Condition Code Control. 4-bit field, one bit of which indicates if the condition code is to be set by the operation.
+  static int64_t vccc_mask(int64_t flag, int pos) {
+    assert((flag == VOP_CCIGN) || (flag == VOP_CCSET), "VCCC flag value out of range");
+    return uimm4(flag, pos, 48);
+  }
+
  public:
 
   //--------------------------------------------------
@@ -1453,6 +1708,8 @@
   static long imm24(int64_t i24, int s, int len)   { return imm(i24, 24) << (len-s-24); }
   static long imm32(int64_t i32, int s, int len)   { return imm(i32, 32) << (len-s-32); }
 
+  static long vreg(VectorRegister v, int pos)      { const int len = 48; return u_field(v->encoding()&0x0f, (len-pos)-1, (len-pos)-4) | v->RXB_mask(pos); }
+
   static long fregt(FloatRegister r, int s, int len) { return freg(r,s,len); }
   static long freg( FloatRegister r, int s, int len) { return u_field(r->encoding(), (len-s)-1, (len-s)-4); }
 
@@ -2125,6 +2382,397 @@
   inline void z_trtt(Register r1, Register r2, int64_t m3);
 
 
+  //---------------------------
+  //--  Vector Instructions  --
+  //---------------------------
+
+  //---<  Vector Support Instructions  >---
+
+  // Load (transfer from memory)
+  inline void z_vlm(   VectorRegister v1, VectorRegister v3, int64_t d2, Register b2);
+  inline void z_vl(    VectorRegister v1, int64_t d2, Register x2, Register b2);
+  inline void z_vleb(  VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3);
+  inline void z_vleh(  VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3);
+  inline void z_vlef(  VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3);
+  inline void z_vleg(  VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3);
+
+  // Gather/Scatter
+  inline void z_vgef(  VectorRegister v1, int64_t d2, VectorRegister vx2, Register b2, int64_t m3);
+  inline void z_vgeg(  VectorRegister v1, int64_t d2, VectorRegister vx2, Register b2, int64_t m3);
+
+  inline void z_vscef( VectorRegister v1, int64_t d2, VectorRegister vx2, Register b2, int64_t m3);
+  inline void z_vsceg( VectorRegister v1, int64_t d2, VectorRegister vx2, Register b2, int64_t m3);
+
+  // load and replicate
+  inline void z_vlrep( VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3);
+  inline void z_vlrepb(VectorRegister v1, int64_t d2, Register x2, Register b2);
+  inline void z_vlreph(VectorRegister v1, int64_t d2, Register x2, Register b2);
+  inline void z_vlrepf(VectorRegister v1, int64_t d2, Register x2, Register b2);
+  inline void z_vlrepg(VectorRegister v1, int64_t d2, Register x2, Register b2);
+
+  inline void z_vllez( VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3);
+  inline void z_vllezb(VectorRegister v1, int64_t d2, Register x2, Register b2);
+  inline void z_vllezh(VectorRegister v1, int64_t d2, Register x2, Register b2);
+  inline void z_vllezf(VectorRegister v1, int64_t d2, Register x2, Register b2);
+  inline void z_vllezg(VectorRegister v1, int64_t d2, Register x2, Register b2);
+
+  inline void z_vlbb(  VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3);
+  inline void z_vll(   VectorRegister v1, Register r3, int64_t d2, Register b2);
+
+  // Load (register to register)
+  inline void z_vlr(   VectorRegister v1, VectorRegister v2);
+
+  inline void z_vlgv(  Register r1, VectorRegister v3, int64_t d2, Register b2, int64_t m4);
+  inline void z_vlgvb( Register r1, VectorRegister v3, int64_t d2, Register b2);
+  inline void z_vlgvh( Register r1, VectorRegister v3, int64_t d2, Register b2);
+  inline void z_vlgvf( Register r1, VectorRegister v3, int64_t d2, Register b2);
+  inline void z_vlgvg( Register r1, VectorRegister v3, int64_t d2, Register b2);
+
+  inline void z_vlvg(  VectorRegister v1, Register r3, int64_t d2, Register b2, int64_t m4);
+  inline void z_vlvgb( VectorRegister v1, Register r3, int64_t d2, Register b2);
+  inline void z_vlvgh( VectorRegister v1, Register r3, int64_t d2, Register b2);
+  inline void z_vlvgf( VectorRegister v1, Register r3, int64_t d2, Register b2);
+  inline void z_vlvgg( VectorRegister v1, Register r3, int64_t d2, Register b2);
+
+  inline void z_vlvgp( VectorRegister v1, Register r2, Register r3);
+
+  // vector register pack
+  inline void z_vpk(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vpkh(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vpkf(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vpkg(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+
+  inline void z_vpks(  VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4, int64_t cc5);
+  inline void z_vpksh( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vpksf( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vpksg( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vpkshs(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vpksfs(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vpksgs(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+
+  inline void z_vpkls(  VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4, int64_t cc5);
+  inline void z_vpklsh( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vpklsf( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vpklsg( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vpklshs(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vpklsfs(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vpklsgs(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+
+  // vector register unpack (sign-extended)
+  inline void z_vuph(   VectorRegister v1, VectorRegister v2, int64_t m3);
+  inline void z_vuphb(  VectorRegister v1, VectorRegister v2);
+  inline void z_vuphh(  VectorRegister v1, VectorRegister v2);
+  inline void z_vuphf(  VectorRegister v1, VectorRegister v2);
+  inline void z_vupl(   VectorRegister v1, VectorRegister v2, int64_t m3);
+  inline void z_vuplb(  VectorRegister v1, VectorRegister v2);
+  inline void z_vuplh(  VectorRegister v1, VectorRegister v2);
+  inline void z_vuplf(  VectorRegister v1, VectorRegister v2);
+
+  // vector register unpack (zero-extended)
+  inline void z_vuplh(  VectorRegister v1, VectorRegister v2, int64_t m3);
+  inline void z_vuplhb( VectorRegister v1, VectorRegister v2);
+  inline void z_vuplhh( VectorRegister v1, VectorRegister v2);
+  inline void z_vuplhf( VectorRegister v1, VectorRegister v2);
+  inline void z_vupll(  VectorRegister v1, VectorRegister v2, int64_t m3);
+  inline void z_vupllb( VectorRegister v1, VectorRegister v2);
+  inline void z_vupllh( VectorRegister v1, VectorRegister v2);
+  inline void z_vupllf( VectorRegister v1, VectorRegister v2);
+
+  // vector register merge high/low
+  inline void z_vmrh( VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vmrhb(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmrhh(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmrhf(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmrhg(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+
+  inline void z_vmrl( VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vmrlb(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmrlh(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmrlf(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmrlg(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+
+  // vector register permute
+  inline void z_vperm( VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4);
+  inline void z_vpdi(  VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t        m4);
+
+  // vector register replicate
+  inline void z_vrep(  VectorRegister v1, VectorRegister v3, int64_t imm2, int64_t m4);
+  inline void z_vrepb( VectorRegister v1, VectorRegister v3, int64_t imm2);
+  inline void z_vreph( VectorRegister v1, VectorRegister v3, int64_t imm2);
+  inline void z_vrepf( VectorRegister v1, VectorRegister v3, int64_t imm2);
+  inline void z_vrepg( VectorRegister v1, VectorRegister v3, int64_t imm2);
+  inline void z_vrepi( VectorRegister v1, int64_t imm2,      int64_t m3);
+  inline void z_vrepib(VectorRegister v1, int64_t imm2);
+  inline void z_vrepih(VectorRegister v1, int64_t imm2);
+  inline void z_vrepif(VectorRegister v1, int64_t imm2);
+  inline void z_vrepig(VectorRegister v1, int64_t imm2);
+
+  inline void z_vsel(  VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4);
+  inline void z_vseg(  VectorRegister v1, VectorRegister v2, int64_t imm3);
+
+  // Load (immediate)
+  inline void z_vleib( VectorRegister v1, int64_t imm2, int64_t m3);
+  inline void z_vleih( VectorRegister v1, int64_t imm2, int64_t m3);
+  inline void z_vleif( VectorRegister v1, int64_t imm2, int64_t m3);
+  inline void z_vleig( VectorRegister v1, int64_t imm2, int64_t m3);
+
+  // Store
+  inline void z_vstm(  VectorRegister v1, VectorRegister v3, int64_t d2, Register b2);
+  inline void z_vst(   VectorRegister v1, int64_t d2, Register x2, Register b2);
+  inline void z_vsteb( VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3);
+  inline void z_vsteh( VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3);
+  inline void z_vstef( VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3);
+  inline void z_vsteg( VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3);
+  inline void z_vstl(  VectorRegister v1, Register r3, int64_t d2, Register b2);
+
+  // Misc
+  inline void z_vgm(   VectorRegister v1, int64_t imm2, int64_t imm3, int64_t m4);
+  inline void z_vgmb(  VectorRegister v1, int64_t imm2, int64_t imm3);
+  inline void z_vgmh(  VectorRegister v1, int64_t imm2, int64_t imm3);
+  inline void z_vgmf(  VectorRegister v1, int64_t imm2, int64_t imm3);
+  inline void z_vgmg(  VectorRegister v1, int64_t imm2, int64_t imm3);
+
+  inline void z_vgbm(  VectorRegister v1, int64_t imm2);
+  inline void z_vzero( VectorRegister v1); // preferred method to set vreg to all zeroes
+  inline void z_vone(  VectorRegister v1); // preferred method to set vreg to all ones
+
+  //---<  Vector Arithmetic Instructions  >---
+
+  // Load
+  inline void z_vlc(    VectorRegister v1, VectorRegister v2, int64_t m3);
+  inline void z_vlcb(   VectorRegister v1, VectorRegister v2);
+  inline void z_vlch(   VectorRegister v1, VectorRegister v2);
+  inline void z_vlcf(   VectorRegister v1, VectorRegister v2);
+  inline void z_vlcg(   VectorRegister v1, VectorRegister v2);
+  inline void z_vlp(    VectorRegister v1, VectorRegister v2, int64_t m3);
+  inline void z_vlpb(   VectorRegister v1, VectorRegister v2);
+  inline void z_vlph(   VectorRegister v1, VectorRegister v2);
+  inline void z_vlpf(   VectorRegister v1, VectorRegister v2);
+  inline void z_vlpg(   VectorRegister v1, VectorRegister v2);
+
+  // ADD
+  inline void z_va(     VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vab(    VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vah(    VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vaf(    VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vag(    VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vaq(    VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vacc(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vaccb(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vacch(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vaccf(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vaccg(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vaccq(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+
+  // SUB
+  inline void z_vs(     VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vsb(    VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vsh(    VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vsf(    VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vsg(    VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vsq(    VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vscbi(  VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vscbib( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vscbih( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vscbif( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vscbig( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vscbiq( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+
+  // MULTIPLY
+  inline void z_vml(    VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vmh(    VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vmlh(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vme(    VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vmle(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vmo(    VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vmlo(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+
+  // MULTIPLY & ADD
+  inline void z_vmal(   VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4, int64_t m5);
+  inline void z_vmah(   VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4, int64_t m5);
+  inline void z_vmalh(  VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4, int64_t m5);
+  inline void z_vmae(   VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4, int64_t m5);
+  inline void z_vmale(  VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4, int64_t m5);
+  inline void z_vmao(   VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4, int64_t m5);
+  inline void z_vmalo(  VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4, int64_t m5);
+
+  // VECTOR SUM
+  inline void z_vsum(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vsumb(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vsumh(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vsumg(  VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vsumgh( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vsumgf( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vsumq(  VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vsumqf( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vsumqg( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+
+  // Average
+  inline void z_vavg(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vavgb(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vavgh(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vavgf(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vavgg(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vavgl(  VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vavglb( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vavglh( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vavglf( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vavglg( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+
+  // VECTOR Galois Field Multiply Sum
+  inline void z_vgfm(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vgfmb(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vgfmh(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vgfmf(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vgfmg(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  // VECTOR Galois Field Multiply Sum and Accumulate
+  inline void z_vgfma(  VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4, int64_t m5);
+  inline void z_vgfmab( VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4);
+  inline void z_vgfmah( VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4);
+  inline void z_vgfmaf( VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4);
+  inline void z_vgfmag( VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4);
+
+  //---<  Vector Logical Instructions  >---
+
+  // AND
+  inline void z_vn(     VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vnc(    VectorRegister v1, VectorRegister v2, VectorRegister v3);
+
+  // XOR
+  inline void z_vx(     VectorRegister v1, VectorRegister v2, VectorRegister v3);
+
+  // NOR
+  inline void z_vno(    VectorRegister v1, VectorRegister v2, VectorRegister v3);
+
+  // OR
+  inline void z_vo(     VectorRegister v1, VectorRegister v2, VectorRegister v3);
+
+  // Comparison (element-wise)
+  inline void z_vceq(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4, int64_t cc5);
+  inline void z_vceqb(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vceqh(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vceqf(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vceqg(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vceqbs( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vceqhs( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vceqfs( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vceqgs( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vch(    VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4, int64_t cc5);
+  inline void z_vchb(   VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vchh(   VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vchf(   VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vchg(   VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vchbs(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vchhs(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vchfs(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vchgs(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vchl(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4, int64_t cc5);
+  inline void z_vchlb(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vchlh(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vchlf(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vchlg(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vchlbs( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vchlhs( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vchlfs( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vchlgs( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+
+  // Max/Min (element-wise)
+  inline void z_vmx(    VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vmxb(   VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmxh(   VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmxf(   VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmxg(   VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmxl(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vmxlb(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmxlh(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmxlf(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmxlg(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmn(    VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vmnb(   VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmnh(   VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmnf(   VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmng(   VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmnl(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4);
+  inline void z_vmnlb(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmnlh(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmnlf(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vmnlg(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+
+  // Leading/Trailing Zeros, population count
+  inline void z_vclz(   VectorRegister v1, VectorRegister v2, int64_t m3);
+  inline void z_vclzb(  VectorRegister v1, VectorRegister v2);
+  inline void z_vclzh(  VectorRegister v1, VectorRegister v2);
+  inline void z_vclzf(  VectorRegister v1, VectorRegister v2);
+  inline void z_vclzg(  VectorRegister v1, VectorRegister v2);
+  inline void z_vctz(   VectorRegister v1, VectorRegister v2, int64_t m3);
+  inline void z_vctzb(  VectorRegister v1, VectorRegister v2);
+  inline void z_vctzh(  VectorRegister v1, VectorRegister v2);
+  inline void z_vctzf(  VectorRegister v1, VectorRegister v2);
+  inline void z_vctzg(  VectorRegister v1, VectorRegister v2);
+  inline void z_vpopct( VectorRegister v1, VectorRegister v2, int64_t m3);
+
+  // Rotate/Shift
+  inline void z_verllv( VectorRegister v1, VectorRegister v2, VectorRegister v3,               int64_t m4);
+  inline void z_verllvb(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_verllvh(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_verllvf(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_verllvg(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_verll(  VectorRegister v1, VectorRegister v3, int64_t d2, Register b2,         int64_t m4);
+  inline void z_verllb( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2);
+  inline void z_verllh( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2);
+  inline void z_verllf( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2);
+  inline void z_verllg( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2);
+  inline void z_verim(  VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t imm4, int64_t m5);
+  inline void z_verimb( VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t imm4);
+  inline void z_verimh( VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t imm4);
+  inline void z_verimf( VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t imm4);
+  inline void z_verimg( VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t imm4);
+
+  inline void z_veslv(  VectorRegister v1, VectorRegister v2, VectorRegister v3,               int64_t m4);
+  inline void z_veslvb( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_veslvh( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_veslvf( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_veslvg( VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vesl(   VectorRegister v1, VectorRegister v3, int64_t d2, Register b2,         int64_t m4);
+  inline void z_veslb(  VectorRegister v1, VectorRegister v3, int64_t d2, Register b2);
+  inline void z_veslh(  VectorRegister v1, VectorRegister v3, int64_t d2, Register b2);
+  inline void z_veslf(  VectorRegister v1, VectorRegister v3, int64_t d2, Register b2);
+  inline void z_veslg(  VectorRegister v1, VectorRegister v3, int64_t d2, Register b2);
+
+  inline void z_vesrav( VectorRegister v1, VectorRegister v2, VectorRegister v3,               int64_t m4);
+  inline void z_vesravb(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vesravh(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vesravf(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vesravg(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vesra(  VectorRegister v1, VectorRegister v3, int64_t d2, Register b2,         int64_t m4);
+  inline void z_vesrab( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2);
+  inline void z_vesrah( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2);
+  inline void z_vesraf( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2);
+  inline void z_vesrag( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2);
+  inline void z_vesrlv( VectorRegister v1, VectorRegister v2, VectorRegister v3,               int64_t m4);
+  inline void z_vesrlvb(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vesrlvh(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vesrlvf(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vesrlvg(VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vesrl(  VectorRegister v1, VectorRegister v3, int64_t d2, Register b2,         int64_t m4);
+  inline void z_vesrlb( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2);
+  inline void z_vesrlh( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2);
+  inline void z_vesrlf( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2);
+  inline void z_vesrlg( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2);
+
+  inline void z_vsl(    VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vslb(   VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vsldb(  VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t imm4);
+
+  inline void z_vsra(   VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vsrab(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vsrl(   VectorRegister v1, VectorRegister v2, VectorRegister v3);
+  inline void z_vsrlb(  VectorRegister v1, VectorRegister v2, VectorRegister v3);
+
+  // Test under Mask
+  inline void z_vtm(    VectorRegister v1, VectorRegister v2);
+
+
   // Floatingpoint instructions
   // ==========================
 
--- a/src/hotspot/cpu/s390/assembler_s390.inline.hpp	Tue Sep 26 06:53:35 2017 -0400
+++ b/src/hotspot/cpu/s390/assembler_s390.inline.hpp	Tue Sep 26 17:06:06 2017 +0200
@@ -1,6 +1,6 @@
 /*
- * Copyright (c) 2016, Oracle and/or its affiliates. All rights reserved.
- * Copyright (c) 2016 SAP SE. All rights reserved.
+ * Copyright (c) 2016, 2017, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2016, 2017 SAP SE. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -702,6 +702,396 @@
 inline void Assembler::z_cvdg(Register r1, int64_t d2, Register x2, Register b2) { emit_48( CVDG_ZOPC | regt(r1, 8, 48) | reg(x2, 12, 48) | reg(b2, 16, 48) | simm20(d2)); }
 
 
+//---------------------------
+//--  Vector Instructions  --
+//---------------------------
+
+//---<  Vector Support Instructions  >---
+
+// Load (transfer from memory)
+inline void Assembler::z_vlm(    VectorRegister v1, VectorRegister v3, int64_t d2, Register b2)       {emit_48(VLM_ZOPC   | vreg(v1,  8)     | vreg(v3, 12)     | rsmask_48(d2,     b2)); }
+inline void Assembler::z_vl(     VectorRegister v1, int64_t d2, Register x2, Register b2)             {emit_48(VL_ZOPC    | vreg(v1,  8)                        | rxmask_48(d2, x2, b2)); }
+inline void Assembler::z_vleb(   VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3) {emit_48(VLEB_ZOPC  | vreg(v1,  8)                        | rxmask_48(d2, x2, b2) | veix_mask(m3, VRET_BYTE, 32)); }
+inline void Assembler::z_vleh(   VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3) {emit_48(VLEH_ZOPC  | vreg(v1,  8)                        | rxmask_48(d2, x2, b2) | veix_mask(m3, VRET_HW,   32)); }
+inline void Assembler::z_vlef(   VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3) {emit_48(VLEF_ZOPC  | vreg(v1,  8)                        | rxmask_48(d2, x2, b2) | veix_mask(m3, VRET_FW,   32)); }
+inline void Assembler::z_vleg(   VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3) {emit_48(VLEG_ZOPC  | vreg(v1,  8)                        | rxmask_48(d2, x2, b2) | veix_mask(m3, VRET_DW,   32)); }
+
+// Gather/Scatter
+inline void Assembler::z_vgef(   VectorRegister v1, int64_t d2, VectorRegister vx2, Register b2, int64_t m3) {emit_48(VGEF_ZOPC  | vreg(v1,  8)                 | rvmask_48(d2, vx2, b2) | veix_mask(m3, VRET_FW,   32)); }
+inline void Assembler::z_vgeg(   VectorRegister v1, int64_t d2, VectorRegister vx2, Register b2, int64_t m3) {emit_48(VGEG_ZOPC  | vreg(v1,  8)                 | rvmask_48(d2, vx2, b2) | veix_mask(m3, VRET_DW,   32)); }
+
+inline void Assembler::z_vscef(  VectorRegister v1, int64_t d2, VectorRegister vx2, Register b2, int64_t m3) {emit_48(VSCEF_ZOPC | vreg(v1,  8)                 | rvmask_48(d2, vx2, b2) | veix_mask(m3, VRET_FW,   32)); }
+inline void Assembler::z_vsceg(  VectorRegister v1, int64_t d2, VectorRegister vx2, Register b2, int64_t m3) {emit_48(VSCEG_ZOPC | vreg(v1,  8)                 | rvmask_48(d2, vx2, b2) | veix_mask(m3, VRET_DW,   32)); }
+
+// load and replicate
+inline void Assembler::z_vlrep(  VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3) {emit_48(VLREP_ZOPC | vreg(v1,  8)                        | rxmask_48(d2, x2, b2) | vesc_mask(m3, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vlrepb( VectorRegister v1, int64_t d2, Register x2, Register b2)             {z_vlrep(v1, d2, x2, b2, VRET_BYTE); }// load byte and replicate to all vector elements of type 'B'
+inline void Assembler::z_vlreph( VectorRegister v1, int64_t d2, Register x2, Register b2)             {z_vlrep(v1, d2, x2, b2, VRET_HW); }  // load HW   and replicate to all vector elements of type 'H'
+inline void Assembler::z_vlrepf( VectorRegister v1, int64_t d2, Register x2, Register b2)             {z_vlrep(v1, d2, x2, b2, VRET_FW); }  // load FW   and replicate to all vector elements of type 'F'
+inline void Assembler::z_vlrepg( VectorRegister v1, int64_t d2, Register x2, Register b2)             {z_vlrep(v1, d2, x2, b2, VRET_DW); }  // load DW   and replicate to all vector elements of type 'G'
+
+inline void Assembler::z_vllez(  VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3) {emit_48(VLLEZ_ZOPC | vreg(v1,  8)                        | rxmask_48(d2, x2, b2) | vesc_mask(m3, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vllezb( VectorRegister v1, int64_t d2, Register x2, Register b2)             {z_vllez(v1, d2, x2, b2, VRET_BYTE); }// load logical byte into left DW of VR, zero all other bit positions.
+inline void Assembler::z_vllezh( VectorRegister v1, int64_t d2, Register x2, Register b2)             {z_vllez(v1, d2, x2, b2, VRET_HW); }  // load logical HW   into left DW of VR, zero all other bit positions.
+inline void Assembler::z_vllezf( VectorRegister v1, int64_t d2, Register x2, Register b2)             {z_vllez(v1, d2, x2, b2, VRET_FW); }  // load logical FW   into left DW of VR, zero all other bit positions.
+inline void Assembler::z_vllezg( VectorRegister v1, int64_t d2, Register x2, Register b2)             {z_vllez(v1, d2, x2, b2, VRET_DW); }  // load logical DW   into left DW of VR, zero all other bit positions.
+
+inline void Assembler::z_vlbb(   VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3) {emit_48(VLBB_ZOPC  | vreg(v1,  8)                        | rxmask_48(d2, x2, b2) | uimm4(m3, 32, 48)); }
+inline void Assembler::z_vll(    VectorRegister v1, Register r3, int64_t d2, Register b2)             {emit_48(VLL_ZOPC   | vreg(v1,  8)     |  reg(r3, 12, 48) | rsmask_48(d2,     b2)); }
+
+// Load (register to register)
+inline void Assembler::z_vlr (   VectorRegister v1, VectorRegister v2)                                {emit_48(VLR_ZOPC   | vreg(v1,  8)     | vreg(v2, 12)); }
+
+inline void Assembler::z_vlgv(   Register r1, VectorRegister v3, int64_t d2, Register b2, int64_t m4) {emit_48(VLGV_ZOPC  |  reg(r1,  8, 48) | vreg(v3, 12)     | rsmask_48(d2,     b2) | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vlgvb(  Register r1, VectorRegister v3, int64_t d2, Register b2)             {z_vlgv(r1, v3, d2, b2, VRET_BYTE); } // load byte from VR element (index d2(b2)) into GR (logical)
+inline void Assembler::z_vlgvh(  Register r1, VectorRegister v3, int64_t d2, Register b2)             {z_vlgv(r1, v3, d2, b2, VRET_HW); }   // load HW   from VR element (index d2(b2)) into GR (logical)
+inline void Assembler::z_vlgvf(  Register r1, VectorRegister v3, int64_t d2, Register b2)             {z_vlgv(r1, v3, d2, b2, VRET_FW); }   // load FW   from VR element (index d2(b2)) into GR (logical)
+inline void Assembler::z_vlgvg(  Register r1, VectorRegister v3, int64_t d2, Register b2)             {z_vlgv(r1, v3, d2, b2, VRET_DW); }   // load DW   from VR element (index d2(b2)) into GR.
+
+inline void Assembler::z_vlvg(   VectorRegister v1, Register r3, int64_t d2, Register b2, int64_t m4) {emit_48(VLVG_ZOPC  | vreg(v1,  8)     |  reg(r3, 12, 48) | rsmask_48(d2,     b2) | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vlvgb(  VectorRegister v1, Register r3, int64_t d2, Register b2)             {z_vlvg(v1, r3, d2, b2, VRET_BYTE); }
+inline void Assembler::z_vlvgh(  VectorRegister v1, Register r3, int64_t d2, Register b2)             {z_vlvg(v1, r3, d2, b2, VRET_HW); }
+inline void Assembler::z_vlvgf(  VectorRegister v1, Register r3, int64_t d2, Register b2)             {z_vlvg(v1, r3, d2, b2, VRET_FW); }
+inline void Assembler::z_vlvgg(  VectorRegister v1, Register r3, int64_t d2, Register b2)             {z_vlvg(v1, r3, d2, b2, VRET_DW); }
+
+inline void Assembler::z_vlvgp(  VectorRegister v1, Register r2, Register r3)                         {emit_48(VLVGP_ZOPC | vreg(v1,  8)     |  reg(r2, 12, 48) |  reg(r3, 16, 48)); }
+
+// vector register pack
+inline void Assembler::z_vpk(    VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VPK_ZOPC   | vreg(v1,  8)     | vreg(v2, 12)     | vreg(v3, 16)     | vesc_mask(m4, VRET_HW, VRET_DW, 32)); }
+inline void Assembler::z_vpkh(   VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vpk(v1, v2, v3, VRET_HW); }       // vector element type 'H'
+inline void Assembler::z_vpkf(   VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vpk(v1, v2, v3, VRET_FW); }       // vector element type 'F'
+inline void Assembler::z_vpkg(   VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vpk(v1, v2, v3, VRET_DW); }       // vector element type 'G'
+
+inline void Assembler::z_vpks(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4, int64_t cc5) {emit_48(VPKS_ZOPC  | vreg(v1,  8) |  vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_HW, VRET_DW, 32) | vccc_mask(cc5, 24)); }
+inline void Assembler::z_vpksh(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vpks(v1, v2, v3, VRET_HW, VOP_CCIGN); }   // vector element type 'H', don't set CC
+inline void Assembler::z_vpksf(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vpks(v1, v2, v3, VRET_FW, VOP_CCIGN); }   // vector element type 'F', don't set CC
+inline void Assembler::z_vpksg(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vpks(v1, v2, v3, VRET_DW, VOP_CCIGN); }   // vector element type 'G', don't set CC
+inline void Assembler::z_vpkshs( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vpks(v1, v2, v3, VRET_HW, VOP_CCSET); }   // vector element type 'H', set CC
+inline void Assembler::z_vpksfs( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vpks(v1, v2, v3, VRET_FW, VOP_CCSET); }   // vector element type 'F', set CC
+inline void Assembler::z_vpksgs( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vpks(v1, v2, v3, VRET_DW, VOP_CCSET); }   // vector element type 'G', set CC
+
+inline void Assembler::z_vpkls(  VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4, int64_t cc5) {emit_48(VPKLS_ZOPC | vreg(v1,  8) |  vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_HW, VRET_DW, 32) | vccc_mask(cc5, 24)); }
+inline void Assembler::z_vpklsh( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vpkls(v1, v2, v3, VRET_HW, VOP_CCIGN); }  // vector element type 'H', don't set CC
+inline void Assembler::z_vpklsf( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vpkls(v1, v2, v3, VRET_FW, VOP_CCIGN); }  // vector element type 'F', don't set CC
+inline void Assembler::z_vpklsg( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vpkls(v1, v2, v3, VRET_DW, VOP_CCIGN); }  // vector element type 'G', don't set CC
+inline void Assembler::z_vpklshs(VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vpkls(v1, v2, v3, VRET_HW, VOP_CCSET); }  // vector element type 'H', set CC
+inline void Assembler::z_vpklsfs(VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vpkls(v1, v2, v3, VRET_FW, VOP_CCSET); }  // vector element type 'F', set CC
+inline void Assembler::z_vpklsgs(VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vpkls(v1, v2, v3, VRET_DW, VOP_CCSET); }  // vector element type 'G', set CC
+
+// vector register unpack (sign-extended)
+inline void Assembler::z_vuph(   VectorRegister v1, VectorRegister v2, int64_t m3)                    {emit_48(VUPH_ZOPC  | vreg(v1,  8)     | vreg(v2, 12)     | vesc_mask(m3, VRET_BYTE, VRET_FW, 32)); }
+inline void Assembler::z_vuphb(  VectorRegister v1, VectorRegister v2)                                {z_vuph(v1, v2, VRET_BYTE); }        // vector element type 'B'
+inline void Assembler::z_vuphh(  VectorRegister v1, VectorRegister v2)                                {z_vuph(v1, v2, VRET_HW); }          // vector element type 'H'
+inline void Assembler::z_vuphf(  VectorRegister v1, VectorRegister v2)                                {z_vuph(v1, v2, VRET_FW); }          // vector element type 'F'
+inline void Assembler::z_vupl(   VectorRegister v1, VectorRegister v2, int64_t m3)                    {emit_48(VUPL_ZOPC  | vreg(v1,  8)     | vreg(v2, 12)     | vesc_mask(m3, VRET_BYTE, VRET_FW, 32)); }
+inline void Assembler::z_vuplb(  VectorRegister v1, VectorRegister v2)                                {z_vupl(v1, v2, VRET_BYTE); }        // vector element type 'B'
+inline void Assembler::z_vuplh(  VectorRegister v1, VectorRegister v2)                                {z_vupl(v1, v2, VRET_HW); }          // vector element type 'H'
+inline void Assembler::z_vuplf(  VectorRegister v1, VectorRegister v2)                                {z_vupl(v1, v2, VRET_FW); }          // vector element type 'F'
+
+// vector register unpack (zero-extended)
+inline void Assembler::z_vuplh(  VectorRegister v1, VectorRegister v2, int64_t m3)                    {emit_48(VUPLH_ZOPC | vreg(v1,  8)     | vreg(v2, 12)     | vesc_mask(m3, VRET_BYTE, VRET_FW, 32)); }
+inline void Assembler::z_vuplhb( VectorRegister v1, VectorRegister v2)                                {z_vuplh(v1, v2, VRET_BYTE); }       // vector element type 'B'
+inline void Assembler::z_vuplhh( VectorRegister v1, VectorRegister v2)                                {z_vuplh(v1, v2, VRET_HW); }         // vector element type 'H'
+inline void Assembler::z_vuplhf( VectorRegister v1, VectorRegister v2)                                {z_vuplh(v1, v2, VRET_FW); }         // vector element type 'F'
+inline void Assembler::z_vupll(  VectorRegister v1, VectorRegister v2, int64_t m3)                    {emit_48(VUPLL_ZOPC | vreg(v1,  8)     | vreg(v2, 12)     | vesc_mask(m3, VRET_BYTE, VRET_FW, 32)); }
+inline void Assembler::z_vupllb( VectorRegister v1, VectorRegister v2)                                {z_vupll(v1, v2, VRET_BYTE); }       // vector element type 'B'
+inline void Assembler::z_vupllh( VectorRegister v1, VectorRegister v2)                                {z_vupll(v1, v2, VRET_HW); }         // vector element type 'H'
+inline void Assembler::z_vupllf( VectorRegister v1, VectorRegister v2)                                {z_vupll(v1, v2, VRET_FW); }         // vector element type 'F'
+
+// vector register merge high/low
+inline void Assembler::z_vmrh(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VMRH_ZOPC  | vreg(v1,  8)     | vreg(v2, 12)     | vreg(v3, 16)     | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vmrhb(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmrh(v1, v2, v3, VRET_BYTE); }    // vector element type 'B'
+inline void Assembler::z_vmrhh(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmrh(v1, v2, v3, VRET_HW); }      // vector element type 'H'
+inline void Assembler::z_vmrhf(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmrh(v1, v2, v3, VRET_FW); }      // vector element type 'F'
+inline void Assembler::z_vmrhg(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmrh(v1, v2, v3, VRET_DW); }      // vector element type 'G'
+
+inline void Assembler::z_vmrl(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VMRL_ZOPC  | vreg(v1,  8)     | vreg(v2, 12)     | vreg(v3, 16)     | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vmrlb(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmrh(v1, v2, v3, VRET_BYTE); }    // vector element type 'B'
+inline void Assembler::z_vmrlh(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmrh(v1, v2, v3, VRET_HW); }      // vector element type 'H'
+inline void Assembler::z_vmrlf(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmrh(v1, v2, v3, VRET_FW); }      // vector element type 'F'
+inline void Assembler::z_vmrlg(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmrh(v1, v2, v3, VRET_DW); }      // vector element type 'G'
+
+// vector register permute
+inline void Assembler::z_vperm(  VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4) {emit_48(VPERM_ZOPC | vreg(v1,  8) |  vreg(v2, 12) | vreg(v3, 16) | vreg(v4, 32)); }
+inline void Assembler::z_vpdi(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t        m4) {emit_48(VPDI_ZOPC  | vreg(v1,  8) |  vreg(v2, 12) | vreg(v3, 16) | uimm4(m4, 32, 48)); }
+
+// vector register replicate
+inline void Assembler::z_vrep(   VectorRegister v1, VectorRegister v3, int64_t imm2, int64_t m4)      {emit_48(VREP_ZOPC  | vreg(v1,  8)     | vreg(v3, 12)     | simm16(imm2, 16, 48) | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vrepb(  VectorRegister v1, VectorRegister v3, int64_t imm2)                  {z_vrep(v1, v3, imm2, VRET_BYTE); }  // vector element type 'B'
+inline void Assembler::z_vreph(  VectorRegister v1, VectorRegister v3, int64_t imm2)                  {z_vrep(v1, v3, imm2, VRET_HW); }    // vector element type 'H'
+inline void Assembler::z_vrepf(  VectorRegister v1, VectorRegister v3, int64_t imm2)                  {z_vrep(v1, v3, imm2, VRET_FW); }    // vector element type 'F'
+inline void Assembler::z_vrepg(  VectorRegister v1, VectorRegister v3, int64_t imm2)                  {z_vrep(v1, v3, imm2, VRET_DW); }    // vector element type 'G'
+inline void Assembler::z_vrepi(  VectorRegister v1, int64_t imm2,      int64_t m3)                    {emit_48(VREPI_ZOPC | vreg(v1,  8)                        | simm16(imm2, 16, 48) | vesc_mask(m3, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vrepib( VectorRegister v1, int64_t imm2)                                     {z_vrepi(v1, imm2, VRET_BYTE); }     // vector element type 'B'
+inline void Assembler::z_vrepih( VectorRegister v1, int64_t imm2)                                     {z_vrepi(v1, imm2, VRET_HW); }       // vector element type 'B'
+inline void Assembler::z_vrepif( VectorRegister v1, int64_t imm2)                                     {z_vrepi(v1, imm2, VRET_FW); }       // vector element type 'B'
+inline void Assembler::z_vrepig( VectorRegister v1, int64_t imm2)                                     {z_vrepi(v1, imm2, VRET_DW); }       // vector element type 'B'
+
+inline void Assembler::z_vsel(   VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4) {emit_48(VSEL_ZOPC  | vreg(v1,  8) |  vreg(v2, 12) |  vreg(v3, 16) |  vreg(v4, 32)); }
+inline void Assembler::z_vseg(   VectorRegister v1, VectorRegister v2, int64_t m3)                    {emit_48(VSEG_ZOPC  | vreg(v1,  8)     | vreg(v2, 12)     | uimm4(m3, 32, 48)); }
+
+// Load (immediate)
+inline void Assembler::z_vleib(  VectorRegister v1, int64_t imm2, int64_t m3)                         {emit_48(VLEIB_ZOPC | vreg(v1,  8)                        | simm16(imm2, 32, 48)  | veix_mask(m3, VRET_BYTE, 32)); }
+inline void Assembler::z_vleih(  VectorRegister v1, int64_t imm2, int64_t m3)                         {emit_48(VLEIH_ZOPC | vreg(v1,  8)                        | simm16(imm2, 32, 48)  | veix_mask(m3, VRET_HW,   32)); }
+inline void Assembler::z_vleif(  VectorRegister v1, int64_t imm2, int64_t m3)                         {emit_48(VLEIF_ZOPC | vreg(v1,  8)                        | simm16(imm2, 32, 48)  | veix_mask(m3, VRET_FW,   32)); }
+inline void Assembler::z_vleig(  VectorRegister v1, int64_t imm2, int64_t m3)                         {emit_48(VLEIG_ZOPC | vreg(v1,  8)                        | simm16(imm2, 32, 48)  | veix_mask(m3, VRET_DW,   32)); }
+
+// Store
+inline void Assembler::z_vstm(   VectorRegister v1, VectorRegister v3, int64_t d2, Register b2)       {emit_48(VSTM_ZOPC  | vreg(v1,  8)     | vreg(v3, 12)     | rsmask_48(d2,     b2)); }
+inline void Assembler::z_vst(    VectorRegister v1, int64_t d2, Register x2, Register b2)             {emit_48(VST_ZOPC   | vreg(v1,  8)                        | rxmask_48(d2, x2, b2)); }
+inline void Assembler::z_vsteb(  VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3) {emit_48(VSTEB_ZOPC | vreg(v1,  8)                        | rxmask_48(d2, x2, b2) | veix_mask(m3, VRET_BYTE, 32)); }
+inline void Assembler::z_vsteh(  VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3) {emit_48(VSTEH_ZOPC | vreg(v1,  8)                        | rxmask_48(d2, x2, b2) | veix_mask(m3, VRET_HW,   32)); }
+inline void Assembler::z_vstef(  VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3) {emit_48(VSTEF_ZOPC | vreg(v1,  8)                        | rxmask_48(d2, x2, b2) | veix_mask(m3, VRET_FW,   32)); }
+inline void Assembler::z_vsteg(  VectorRegister v1, int64_t d2, Register x2, Register b2, int64_t m3) {emit_48(VSTEG_ZOPC | vreg(v1,  8)                        | rxmask_48(d2, x2, b2) | veix_mask(m3, VRET_DW,   32)); }
+inline void Assembler::z_vstl(   VectorRegister v1, Register r3, int64_t d2, Register b2)             {emit_48(VSTL_ZOPC  | vreg(v1,  8)     |  reg(r3, 12, 48) | rsmask_48(d2,     b2)); }
+
+// Misc
+inline void Assembler::z_vgm(    VectorRegister v1, int64_t imm2, int64_t imm3, int64_t m4)           {emit_48(VGM_ZOPC   | vreg(v1,  8)     | uimm8( imm2, 16, 48) | uimm8(imm3, 24, 48) | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vgmb(   VectorRegister v1, int64_t imm2, int64_t imm3)                       {z_vgm(v1, imm2, imm3, VRET_BYTE); } // vector element type 'B'
+inline void Assembler::z_vgmh(   VectorRegister v1, int64_t imm2, int64_t imm3)                       {z_vgm(v1, imm2, imm3, VRET_HW); }   // vector element type 'H'
+inline void Assembler::z_vgmf(   VectorRegister v1, int64_t imm2, int64_t imm3)                       {z_vgm(v1, imm2, imm3, VRET_FW); }   // vector element type 'F'
+inline void Assembler::z_vgmg(   VectorRegister v1, int64_t imm2, int64_t imm3)                       {z_vgm(v1, imm2, imm3, VRET_DW); }   // vector element type 'G'
+
+inline void Assembler::z_vgbm(   VectorRegister v1, int64_t imm2)                                     {emit_48(VGBM_ZOPC  | vreg(v1,  8)     | uimm16(imm2, 16, 48)); }
+inline void Assembler::z_vzero(  VectorRegister v1)                                                   {z_vgbm(v1, 0); }      // preferred method to set vreg to all zeroes
+inline void Assembler::z_vone(   VectorRegister v1)                                                   {z_vgbm(v1, 0xffff); } // preferred method to set vreg to all ones
+
+//---<  Vector Arithmetic Instructions  >---
+
+// Load
+inline void Assembler::z_vlc(    VectorRegister v1, VectorRegister v2, int64_t m3)                    {emit_48(VLC_ZOPC   | vreg(v1,  8) | vreg(v2, 12) | vesc_mask(m3, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vlcb(   VectorRegister v1, VectorRegister v2)                                {z_vlc(v1, v2, VRET_BYTE); }         // vector element type 'B'
+inline void Assembler::z_vlch(   VectorRegister v1, VectorRegister v2)                                {z_vlc(v1, v2, VRET_HW); }           // vector element type 'H'
+inline void Assembler::z_vlcf(   VectorRegister v1, VectorRegister v2)                                {z_vlc(v1, v2, VRET_FW); }           // vector element type 'F'
+inline void Assembler::z_vlcg(   VectorRegister v1, VectorRegister v2)                                {z_vlc(v1, v2, VRET_DW); }           // vector element type 'G'
+inline void Assembler::z_vlp(    VectorRegister v1, VectorRegister v2, int64_t m3)                    {emit_48(VLP_ZOPC   | vreg(v1,  8) | vreg(v2, 12) | vesc_mask(m3, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vlpb(   VectorRegister v1, VectorRegister v2)                                {z_vlp(v1, v2, VRET_BYTE); }         // vector element type 'B'
+inline void Assembler::z_vlph(   VectorRegister v1, VectorRegister v2)                                {z_vlp(v1, v2, VRET_HW); }           // vector element type 'H'
+inline void Assembler::z_vlpf(   VectorRegister v1, VectorRegister v2)                                {z_vlp(v1, v2, VRET_FW); }           // vector element type 'F'
+inline void Assembler::z_vlpg(   VectorRegister v1, VectorRegister v2)                                {z_vlp(v1, v2, VRET_DW); }           // vector element type 'G'
+
+// ADD
+inline void Assembler::z_va(     VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VA_ZOPC    | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_QW, 32)); }
+inline void Assembler::z_vab(    VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_va(v1, v2, v3, VRET_BYTE); }      // vector element type 'B'
+inline void Assembler::z_vah(    VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_va(v1, v2, v3, VRET_HW); }        // vector element type 'H'
+inline void Assembler::z_vaf(    VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_va(v1, v2, v3, VRET_FW); }        // vector element type 'F'
+inline void Assembler::z_vag(    VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_va(v1, v2, v3, VRET_DW); }        // vector element type 'G'
+inline void Assembler::z_vaq(    VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_va(v1, v2, v3, VRET_QW); }        // vector element type 'Q'
+inline void Assembler::z_vacc(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VACC_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_QW, 32)); }
+inline void Assembler::z_vaccb(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vacc(v1, v2, v3, VRET_BYTE); }    // vector element type 'B'
+inline void Assembler::z_vacch(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vacc(v1, v2, v3, VRET_HW); }      // vector element type 'H'
+inline void Assembler::z_vaccf(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vacc(v1, v2, v3, VRET_FW); }      // vector element type 'F'
+inline void Assembler::z_vaccg(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vacc(v1, v2, v3, VRET_DW); }      // vector element type 'G'
+inline void Assembler::z_vaccq(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vacc(v1, v2, v3, VRET_QW); }      // vector element type 'Q'
+
+// SUB
+inline void Assembler::z_vs(     VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VS_ZOPC    | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_QW, 32)); }
+inline void Assembler::z_vsb(    VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vs(v1, v2, v3, VRET_BYTE); }      // vector element type 'B'
+inline void Assembler::z_vsh(    VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vs(v1, v2, v3, VRET_HW); }        // vector element type 'H'
+inline void Assembler::z_vsf(    VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vs(v1, v2, v3, VRET_FW); }        // vector element type 'F'
+inline void Assembler::z_vsg(    VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vs(v1, v2, v3, VRET_DW); }        // vector element type 'G'
+inline void Assembler::z_vsq(    VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vs(v1, v2, v3, VRET_QW); }        // vector element type 'Q'
+inline void Assembler::z_vscbi(  VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VSCBI_ZOPC | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_QW, 32)); }
+inline void Assembler::z_vscbib( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vscbi(v1, v2, v3, VRET_BYTE); }   // vector element type 'B'
+inline void Assembler::z_vscbih( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vscbi(v1, v2, v3, VRET_HW); }     // vector element type 'H'
+inline void Assembler::z_vscbif( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vscbi(v1, v2, v3, VRET_FW); }     // vector element type 'F'
+inline void Assembler::z_vscbig( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vscbi(v1, v2, v3, VRET_DW); }     // vector element type 'G'
+inline void Assembler::z_vscbiq( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vscbi(v1, v2, v3, VRET_QW); }     // vector element type 'Q'
+
+// MULTIPLY
+inline void Assembler::z_vml(    VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VML_ZOPC   | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_FW, 32)); }
+inline void Assembler::z_vmh(    VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VMH_ZOPC   | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_FW, 32)); }
+inline void Assembler::z_vmlh(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VMLH_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_FW, 32)); }
+inline void Assembler::z_vme(    VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VME_ZOPC   | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_FW, 32)); }
+inline void Assembler::z_vmle(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VMLE_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_FW, 32)); }
+inline void Assembler::z_vmo(    VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VMO_ZOPC   | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_FW, 32)); }
+inline void Assembler::z_vmlo(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VMLO_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_FW, 32)); }
+
+// MULTIPLY & ADD
+inline void Assembler::z_vmal(   VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4, int64_t m5) {emit_48(VMAL_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vreg(v4, 32) | vesc_mask(m5, VRET_BYTE, VRET_FW, 20)); }
+inline void Assembler::z_vmah(   VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4, int64_t m5) {emit_48(VMAH_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vreg(v4, 32) | vesc_mask(m5, VRET_BYTE, VRET_FW, 20)); }
+inline void Assembler::z_vmalh(  VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4, int64_t m5) {emit_48(VMALH_ZOPC | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vreg(v4, 32) | vesc_mask(m5, VRET_BYTE, VRET_FW, 20)); }
+inline void Assembler::z_vmae(   VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4, int64_t m5) {emit_48(VMAE_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vreg(v4, 32) | vesc_mask(m5, VRET_BYTE, VRET_FW, 20)); }
+inline void Assembler::z_vmale(  VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4, int64_t m5) {emit_48(VMALE_ZOPC | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vreg(v4, 32) | vesc_mask(m5, VRET_BYTE, VRET_FW, 20)); }
+inline void Assembler::z_vmao(   VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4, int64_t m5) {emit_48(VMAO_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vreg(v4, 32) | vesc_mask(m5, VRET_BYTE, VRET_FW, 20)); }
+inline void Assembler::z_vmalo(  VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4, int64_t m5) {emit_48(VMALO_ZOPC | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vreg(v4, 32) | vesc_mask(m5, VRET_BYTE, VRET_FW, 20)); }
+
+// VECTOR SUM
+inline void Assembler::z_vsum(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VSUM_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_HW, 32)); }
+inline void Assembler::z_vsumb(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vsum(v1, v2, v3, VRET_BYTE); }    // vector element type 'B'
+inline void Assembler::z_vsumh(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vsum(v1, v2, v3, VRET_HW); }      // vector element type 'H'
+inline void Assembler::z_vsumg(  VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VSUMG_ZOPC | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_HW,   VRET_FW, 32)); }
+inline void Assembler::z_vsumgh( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vsumg(v1, v2, v3, VRET_HW); }     // vector element type 'B'
+inline void Assembler::z_vsumgf( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vsumg(v1, v2, v3, VRET_FW); }     // vector element type 'H'
+inline void Assembler::z_vsumq(  VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VSUMQ_ZOPC | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_FW,   VRET_DW, 32)); }
+inline void Assembler::z_vsumqf( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vsumq(v1, v2, v3, VRET_FW); }     // vector element type 'B'
+inline void Assembler::z_vsumqg( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vsumq(v1, v2, v3, VRET_DW); }     // vector element type 'H'
+
+// Average
+inline void Assembler::z_vavg(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VAVG_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vavgb(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vavg(v1, v2, v3, VRET_BYTE); }    // vector element type 'B'
+inline void Assembler::z_vavgh(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vavg(v1, v2, v3, VRET_HW); }      // vector element type 'H'
+inline void Assembler::z_vavgf(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vavg(v1, v2, v3, VRET_FW); }      // vector element type 'F'
+inline void Assembler::z_vavgg(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vavg(v1, v2, v3, VRET_DW); }      // vector element type 'G'
+inline void Assembler::z_vavgl(  VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VAVGL_ZOPC | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vavglb( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vavgl(v1, v2, v3, VRET_BYTE); }   // vector element type 'B'
+inline void Assembler::z_vavglh( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vavgl(v1, v2, v3, VRET_HW); }     // vector element type 'H'
+inline void Assembler::z_vavglf( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vavgl(v1, v2, v3, VRET_FW); }     // vector element type 'F'
+inline void Assembler::z_vavglg( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vavgl(v1, v2, v3, VRET_DW); }     // vector element type 'G'
+
+// VECTOR Galois Field Multiply Sum
+inline void Assembler::z_vgfm(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VGFM_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vgfmb(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vgfm(v1, v2, v3, VRET_BYTE); }    // vector element type 'B'
+inline void Assembler::z_vgfmh(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vgfm(v1, v2, v3, VRET_HW); }      // vector element type 'H'
+inline void Assembler::z_vgfmf(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vgfm(v1, v2, v3, VRET_FW); }      // vector element type 'F'
+inline void Assembler::z_vgfmg(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vgfm(v1, v2, v3, VRET_DW); }      // vector element type 'G'
+inline void Assembler::z_vgfma(  VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4, int64_t m5) {emit_48(VGFMA_ZOPC | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vreg(v3, 16) | vesc_mask(m5, VRET_BYTE, VRET_DW, 20)); }
+inline void Assembler::z_vgfmab( VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4) {z_vgfma(v1, v2, v3, v4, VRET_BYTE); } // vector element type 'B'
+inline void Assembler::z_vgfmah( VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4) {z_vgfma(v1, v2, v3, v4, VRET_HW); }   // vector element type 'H'
+inline void Assembler::z_vgfmaf( VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4) {z_vgfma(v1, v2, v3, v4, VRET_FW); }   // vector element type 'F'
+inline void Assembler::z_vgfmag( VectorRegister v1, VectorRegister v2, VectorRegister v3, VectorRegister v4) {z_vgfma(v1, v2, v3, v4, VRET_DW); }   // vector element type 'G'
+
+//---<  Vector Logical Instructions  >---
+
+// AND
+inline void Assembler::z_vn(     VectorRegister v1, VectorRegister v2, VectorRegister v3)             {emit_48(VN_ZOPC    | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16)); }
+inline void Assembler::z_vnc(    VectorRegister v1, VectorRegister v2, VectorRegister v3)             {emit_48(VNC_ZOPC   | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16)); }
+
+// XOR
+inline void Assembler::z_vx(     VectorRegister v1, VectorRegister v2, VectorRegister v3)             {emit_48(VX_ZOPC    | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16)); }
+
+// NOR
+inline void Assembler::z_vno(    VectorRegister v1, VectorRegister v2, VectorRegister v3)             {emit_48(VNO_ZOPC   | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16)); }
+
+// OR
+inline void Assembler::z_vo(     VectorRegister v1, VectorRegister v2, VectorRegister v3)             {emit_48(VO_ZOPC    | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16)); }
+
+// Comparison (element-wise)
+inline void Assembler::z_vceq(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4, int64_t cc5) {emit_48(VCEQ_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_DW, 32) | vccc_mask(cc5, 24)); }
+inline void Assembler::z_vceqb(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vceq(v1, v2, v3, VRET_BYTE, VOP_CCIGN); } // vector element type 'B', don't set CC
+inline void Assembler::z_vceqh(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vceq(v1, v2, v3, VRET_HW,   VOP_CCIGN); } // vector element type 'H', don't set CC
+inline void Assembler::z_vceqf(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vceq(v1, v2, v3, VRET_FW,   VOP_CCIGN); } // vector element type 'F', don't set CC
+inline void Assembler::z_vceqg(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vceq(v1, v2, v3, VRET_DW,   VOP_CCIGN); } // vector element type 'G', don't set CC
+inline void Assembler::z_vceqbs( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vceq(v1, v2, v3, VRET_BYTE, VOP_CCSET); } // vector element type 'B', don't set CC
+inline void Assembler::z_vceqhs( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vceq(v1, v2, v3, VRET_HW,   VOP_CCSET); } // vector element type 'H', don't set CC
+inline void Assembler::z_vceqfs( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vceq(v1, v2, v3, VRET_FW,   VOP_CCSET); } // vector element type 'F', don't set CC
+inline void Assembler::z_vceqgs( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vceq(v1, v2, v3, VRET_DW,   VOP_CCSET); } // vector element type 'G', don't set CC
+inline void Assembler::z_vch(    VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4, int64_t cc5) {emit_48(VCH_ZOPC   | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_DW, 32) | vccc_mask(cc5, 24)); }
+inline void Assembler::z_vchb(   VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vch(v1, v2, v3, VRET_BYTE,  VOP_CCIGN); }  // vector element type 'B', don't set CC
+inline void Assembler::z_vchh(   VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vch(v1, v2, v3, VRET_HW,    VOP_CCIGN); }  // vector element type 'H', don't set CC
+inline void Assembler::z_vchf(   VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vch(v1, v2, v3, VRET_FW,    VOP_CCIGN); }  // vector element type 'F', don't set CC
+inline void Assembler::z_vchg(   VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vch(v1, v2, v3, VRET_DW,    VOP_CCIGN); }  // vector element type 'G', don't set CC
+inline void Assembler::z_vchbs(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vch(v1, v2, v3, VRET_BYTE,  VOP_CCSET); }  // vector element type 'B', don't set CC
+inline void Assembler::z_vchhs(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vch(v1, v2, v3, VRET_HW,    VOP_CCSET); }  // vector element type 'H', don't set CC
+inline void Assembler::z_vchfs(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vch(v1, v2, v3, VRET_FW,    VOP_CCSET); }  // vector element type 'F', don't set CC
+inline void Assembler::z_vchgs(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vch(v1, v2, v3, VRET_DW,    VOP_CCSET); }  // vector element type 'G', don't set CC
+inline void Assembler::z_vchl(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4, int64_t cc5) {emit_48(VCHL_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_DW, 32) | vccc_mask(cc5, 24)); }
+inline void Assembler::z_vchlb(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vchl(v1, v2, v3, VRET_BYTE, VOP_CCIGN); }  // vector element type 'B', don't set CC
+inline void Assembler::z_vchlh(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vchl(v1, v2, v3, VRET_HW,   VOP_CCIGN); }  // vector element type 'H', don't set CC
+inline void Assembler::z_vchlf(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vchl(v1, v2, v3, VRET_FW,   VOP_CCIGN); }  // vector element type 'F', don't set CC
+inline void Assembler::z_vchlg(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vchl(v1, v2, v3, VRET_DW,   VOP_CCIGN); }  // vector element type 'G', don't set CC
+inline void Assembler::z_vchlbs( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vchl(v1, v2, v3, VRET_BYTE, VOP_CCSET); }  // vector element type 'B', don't set CC
+inline void Assembler::z_vchlhs( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vchl(v1, v2, v3, VRET_HW,   VOP_CCSET); }  // vector element type 'H', don't set CC
+inline void Assembler::z_vchlfs( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vchl(v1, v2, v3, VRET_FW,   VOP_CCSET); }  // vector element type 'F', don't set CC
+inline void Assembler::z_vchlgs( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vchl(v1, v2, v3, VRET_DW,   VOP_CCSET); }  // vector element type 'G', don't set CC
+
+// Max/Min (element-wise)
+inline void Assembler::z_vmx(    VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VMX_ZOPC   | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vmxb(   VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmx(v1, v2, v3, VRET_BYTE); }     // vector element type 'B'
+inline void Assembler::z_vmxh(   VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmx(v1, v2, v3, VRET_HW); }       // vector element type 'H'
+inline void Assembler::z_vmxf(   VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmx(v1, v2, v3, VRET_FW); }       // vector element type 'F'
+inline void Assembler::z_vmxg(   VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmx(v1, v2, v3, VRET_DW); }       // vector element type 'G'
+inline void Assembler::z_vmxl(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VMXL_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vmxlb(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmxl(v1, v2, v3, VRET_BYTE); }    // vector element type 'B'
+inline void Assembler::z_vmxlh(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmxl(v1, v2, v3, VRET_HW); }      // vector element type 'H'
+inline void Assembler::z_vmxlf(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmxl(v1, v2, v3, VRET_FW); }      // vector element type 'F'
+inline void Assembler::z_vmxlg(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmxl(v1, v2, v3, VRET_DW); }      // vector element type 'G'
+inline void Assembler::z_vmn(    VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VMN_ZOPC   | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vmnb(   VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmn(v1, v2, v3, VRET_BYTE); }     // vector element type 'B'
+inline void Assembler::z_vmnh(   VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmn(v1, v2, v3, VRET_HW); }       // vector element type 'H'
+inline void Assembler::z_vmnf(   VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmn(v1, v2, v3, VRET_FW); }       // vector element type 'F'
+inline void Assembler::z_vmng(   VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmn(v1, v2, v3, VRET_DW); }       // vector element type 'G'
+inline void Assembler::z_vmnl(   VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t m4) {emit_48(VMNL_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vmnlb(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmnl(v1, v2, v3, VRET_BYTE); }    // vector element type 'B'
+inline void Assembler::z_vmnlh(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmnl(v1, v2, v3, VRET_HW); }      // vector element type 'H'
+inline void Assembler::z_vmnlf(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmnl(v1, v2, v3, VRET_FW); }      // vector element type 'F'
+inline void Assembler::z_vmnlg(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vmnl(v1, v2, v3, VRET_DW); }      // vector element type 'G'
+
+// Leading/Trailing Zeros, population count
+inline void Assembler::z_vclz(   VectorRegister v1, VectorRegister v2, int64_t m3)                    {emit_48(VCLZ_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vesc_mask(m3, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vclzb(  VectorRegister v1, VectorRegister v2)                                {z_vclz(v1, v2, VRET_BYTE); }        // vector element type 'B'
+inline void Assembler::z_vclzh(  VectorRegister v1, VectorRegister v2)                                {z_vclz(v1, v2, VRET_HW); }          // vector element type 'H'
+inline void Assembler::z_vclzf(  VectorRegister v1, VectorRegister v2)                                {z_vclz(v1, v2, VRET_FW); }          // vector element type 'F'
+inline void Assembler::z_vclzg(  VectorRegister v1, VectorRegister v2)                                {z_vclz(v1, v2, VRET_DW); }          // vector element type 'G'
+inline void Assembler::z_vctz(   VectorRegister v1, VectorRegister v2, int64_t m3)                    {emit_48(VCTZ_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vesc_mask(m3, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vctzb(  VectorRegister v1, VectorRegister v2)                                {z_vctz(v1, v2, VRET_BYTE); }        // vector element type 'B'
+inline void Assembler::z_vctzh(  VectorRegister v1, VectorRegister v2)                                {z_vctz(v1, v2, VRET_HW); }          // vector element type 'H'
+inline void Assembler::z_vctzf(  VectorRegister v1, VectorRegister v2)                                {z_vctz(v1, v2, VRET_FW); }          // vector element type 'F'
+inline void Assembler::z_vctzg(  VectorRegister v1, VectorRegister v2)                                {z_vctz(v1, v2, VRET_DW); }          // vector element type 'G'
+inline void Assembler::z_vpopct( VectorRegister v1, VectorRegister v2, int64_t m3)                    {emit_48(VPOPCT_ZOPC| vreg(v1,  8) | vreg(v2, 12) | vesc_mask(m3, VRET_BYTE, VRET_DW, 32)); }
+
+// Rotate/Shift
+inline void Assembler::z_verllv( VectorRegister v1, VectorRegister v2, VectorRegister v3,               int64_t m4) {emit_48(VERLLV_ZOPC| vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16)      | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_verllvb(VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_verllv(v1, v2, v3, VRET_BYTE); }  // vector element type 'B'
+inline void Assembler::z_verllvh(VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_verllv(v1, v2, v3, VRET_HW); }    // vector element type 'H'
+inline void Assembler::z_verllvf(VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_verllv(v1, v2, v3, VRET_FW); }    // vector element type 'F'
+inline void Assembler::z_verllvg(VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_verllv(v1, v2, v3, VRET_DW); }    // vector element type 'G'
+inline void Assembler::z_verll(  VectorRegister v1, VectorRegister v3, int64_t d2, Register b2,         int64_t m4) {emit_48(VERLL_ZOPC | vreg(v1,  8) | vreg(v3, 12) | rsmask_48(d2, b2) | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_verllb( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2)       {z_verll(v1, v3, d2, b2, VRET_BYTE);}// vector element type 'B'
+inline void Assembler::z_verllh( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2)       {z_verll(v1, v3, d2, b2, VRET_HW);}  // vector element type 'H'
+inline void Assembler::z_verllf( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2)       {z_verll(v1, v3, d2, b2, VRET_FW);}  // vector element type 'F'
+inline void Assembler::z_verllg( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2)       {z_verll(v1, v3, d2, b2, VRET_DW);}  // vector element type 'G'
+inline void Assembler::z_verim(  VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t imm4, int64_t m5) {emit_48(VERLL_ZOPC | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16)      | uimm8(imm4, 24, 48) | vesc_mask(m5, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_verimb( VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t imm4) {z_verim(v1, v2, v3, imm4, VRET_BYTE); }   // vector element type 'B'
+inline void Assembler::z_verimh( VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t imm4) {z_verim(v1, v2, v3, imm4, VRET_HW); }     // vector element type 'H'
+inline void Assembler::z_verimf( VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t imm4) {z_verim(v1, v2, v3, imm4, VRET_FW); }     // vector element type 'F'
+inline void Assembler::z_verimg( VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t imm4) {z_verim(v1, v2, v3, imm4, VRET_DW); }     // vector element type 'G'
+
+inline void Assembler::z_veslv(  VectorRegister v1, VectorRegister v2, VectorRegister v3,               int64_t m4) {emit_48(VESLV_ZOPC | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16)      | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_veslvb( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_veslv(v1, v2, v3, VRET_BYTE); }   // vector element type 'B'
+inline void Assembler::z_veslvh( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_veslv(v1, v2, v3, VRET_HW); }     // vector element type 'H'
+inline void Assembler::z_veslvf( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_veslv(v1, v2, v3, VRET_FW); }     // vector element type 'F'
+inline void Assembler::z_veslvg( VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_veslv(v1, v2, v3, VRET_DW); }     // vector element type 'G'
+inline void Assembler::z_vesl(   VectorRegister v1, VectorRegister v3, int64_t d2, Register b2,         int64_t m4) {emit_48(VESL_ZOPC  | vreg(v1,  8) | vreg(v3, 12) | rsmask_48(d2, b2) | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_veslb(  VectorRegister v1, VectorRegister v3, int64_t d2, Register b2)       {z_vesl(v1, v3, d2, b2, VRET_BYTE);} // vector element type 'B'
+inline void Assembler::z_veslh(  VectorRegister v1, VectorRegister v3, int64_t d2, Register b2)       {z_vesl(v1, v3, d2, b2, VRET_HW);}   // vector element type 'H'
+inline void Assembler::z_veslf(  VectorRegister v1, VectorRegister v3, int64_t d2, Register b2)       {z_vesl(v1, v3, d2, b2, VRET_FW);}   // vector element type 'F'
+inline void Assembler::z_veslg(  VectorRegister v1, VectorRegister v3, int64_t d2, Register b2)       {z_vesl(v1, v3, d2, b2, VRET_DW);}   // vector element type 'G'
+
+inline void Assembler::z_vesrav( VectorRegister v1, VectorRegister v2, VectorRegister v3,               int64_t m4) {emit_48(VESRAV_ZOPC| vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16)      | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vesravb(VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vesrav(v1, v2, v3, VRET_BYTE); }  // vector element type 'B'
+inline void Assembler::z_vesravh(VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vesrav(v1, v2, v3, VRET_HW); }    // vector element type 'H'
+inline void Assembler::z_vesravf(VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vesrav(v1, v2, v3, VRET_FW); }    // vector element type 'F'
+inline void Assembler::z_vesravg(VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vesrav(v1, v2, v3, VRET_DW); }    // vector element type 'G'
+inline void Assembler::z_vesra(  VectorRegister v1, VectorRegister v3, int64_t d2, Register b2,         int64_t m4) {emit_48(VESRA_ZOPC | vreg(v1,  8) | vreg(v3, 12) | rsmask_48(d2, b2) | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vesrab( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2)       {z_vesra(v1, v3, d2, b2, VRET_BYTE);}// vector element type 'B'
+inline void Assembler::z_vesrah( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2)       {z_vesra(v1, v3, d2, b2, VRET_HW);}  // vector element type 'H'
+inline void Assembler::z_vesraf( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2)       {z_vesra(v1, v3, d2, b2, VRET_FW);}  // vector element type 'F'
+inline void Assembler::z_vesrag( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2)       {z_vesra(v1, v3, d2, b2, VRET_DW);}  // vector element type 'G'
+inline void Assembler::z_vesrlv( VectorRegister v1, VectorRegister v2, VectorRegister v3,               int64_t m4) {emit_48(VESRLV_ZOPC| vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16)      | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vesrlvb(VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vesrlv(v1, v2, v3, VRET_BYTE); }  // vector element type 'B'
+inline void Assembler::z_vesrlvh(VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vesrlv(v1, v2, v3, VRET_HW); }    // vector element type 'H'
+inline void Assembler::z_vesrlvf(VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vesrlv(v1, v2, v3, VRET_FW); }    // vector element type 'F'
+inline void Assembler::z_vesrlvg(VectorRegister v1, VectorRegister v2, VectorRegister v3)             {z_vesrlv(v1, v2, v3, VRET_DW); }    // vector element type 'G'
+inline void Assembler::z_vesrl(  VectorRegister v1, VectorRegister v3, int64_t d2, Register b2,         int64_t m4) {emit_48(VESRL_ZOPC | vreg(v1,  8) | vreg(v3, 12) | rsmask_48(d2, b2) | vesc_mask(m4, VRET_BYTE, VRET_DW, 32)); }
+inline void Assembler::z_vesrlb( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2)       {z_vesrl(v1, v3, d2, b2, VRET_BYTE);}// vector element type 'B'
+inline void Assembler::z_vesrlh( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2)       {z_vesrl(v1, v3, d2, b2, VRET_HW);}  // vector element type 'H'
+inline void Assembler::z_vesrlf( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2)       {z_vesrl(v1, v3, d2, b2, VRET_FW);}  // vector element type 'F'
+inline void Assembler::z_vesrlg( VectorRegister v1, VectorRegister v3, int64_t d2, Register b2)       {z_vesrl(v1, v3, d2, b2, VRET_DW);}  // vector element type 'G'
+
+inline void Assembler::z_vsl(    VectorRegister v1, VectorRegister v2, VectorRegister v3)               {emit_48(VSL_ZOPC   | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16)); }
+inline void Assembler::z_vslb(   VectorRegister v1, VectorRegister v2, VectorRegister v3)               {emit_48(VSLB_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16)); }
+inline void Assembler::z_vsldb(  VectorRegister v1, VectorRegister v2, VectorRegister v3, int64_t imm4) {emit_48(VSLDB_ZOPC | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16) | uimm8(imm4, 24, 48)); }
+
+inline void Assembler::z_vsra(   VectorRegister v1, VectorRegister v2, VectorRegister v3)             {emit_48(VSRA_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16)); }
+inline void Assembler::z_vsrab(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {emit_48(VSRAB_ZOPC | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16)); }
+inline void Assembler::z_vsrl(   VectorRegister v1, VectorRegister v2, VectorRegister v3)             {emit_48(VSRL_ZOPC  | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16)); }
+inline void Assembler::z_vsrlb(  VectorRegister v1, VectorRegister v2, VectorRegister v3)             {emit_48(VSRLB_ZOPC | vreg(v1,  8) | vreg(v2, 12) | vreg(v3, 16)); }
+
+// Test under Mask
+inline void Assembler::z_vtm(    VectorRegister v1, VectorRegister v2)                                {emit_48(VTM_ZOPC   | vreg(v1,  8) | vreg(v2, 12)); }
+
+
 //-------------------------------
 // FLOAT INSTRUCTIONS
 //-------------------------------
--- a/src/hotspot/cpu/s390/register_definitions_s390.cpp	Tue Sep 26 06:53:35 2017 -0400
+++ b/src/hotspot/cpu/s390/register_definitions_s390.cpp	Tue Sep 26 17:06:06 2017 +0200
@@ -1,6 +1,6 @@
 /*
- * Copyright (c) 2016, Oracle and/or its affiliates. All rights reserved.
- * Copyright (c) 2016 SAP SE. All rights reserved.
+ * Copyright (c) 2016, 2017, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2016, 2017 SAP SE. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -35,3 +35,5 @@
 REGISTER_DEFINITION(Register, noreg);
 
 REGISTER_DEFINITION(FloatRegister, fnoreg);
+
+REGISTER_DEFINITION(VectorRegister, vnoreg);
--- a/src/hotspot/cpu/s390/register_s390.cpp	Tue Sep 26 06:53:35 2017 -0400
+++ b/src/hotspot/cpu/s390/register_s390.cpp	Tue Sep 26 17:06:06 2017 +0200
@@ -1,6 +1,6 @@
 /*
- * Copyright (c) 2016, Oracle and/or its affiliates. All rights reserved.
- * Copyright (c) 2016 SAP SE. All rights reserved.
+ * Copyright (c) 2016, 2017, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2016, 2017 SAP SE. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -46,3 +46,13 @@
   };
   return is_valid() ? names[encoding()] : "fnoreg";
 }
+
+const char* VectorRegisterImpl::name() const {
+  const char* names[number_of_registers] = {
+    "Z_V0",  "Z_V1",  "Z_V2",  "Z_V3",  "Z_V4",  "Z_V5",  "Z_V6",  "Z_V7",
+    "Z_V8",  "Z_V9",  "Z_V10", "Z_V11", "Z_V12", "Z_V13", "Z_V14", "Z_V15",
+    "Z_V16", "Z_V17", "Z_V18", "Z_V19", "Z_V20", "Z_V21", "Z_V22", "Z_V23",
+    "Z_V24", "Z_V25", "Z_V26", "Z_V27", "Z_V28", "Z_V29", "Z_V30", "Z_V31"
+  };
+  return is_valid() ? names[encoding()] : "fnoreg";
+}
--- a/src/hotspot/cpu/s390/register_s390.hpp	Tue Sep 26 06:53:35 2017 -0400
+++ b/src/hotspot/cpu/s390/register_s390.hpp	Tue Sep 26 17:06:06 2017 +0200
@@ -1,6 +1,6 @@
 /*
- * Copyright (c) 2016, Oracle and/or its affiliates. All rights reserved.
- * Copyright (c) 2016 SAP SE. All rights reserved.
+ * Copyright (c) 2016, 2017, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2016, 2017 SAP SE. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -34,11 +34,6 @@
 
 typedef VMRegImpl* VMReg;
 
-// Use Register as shortcut.
-class RegisterImpl;
-typedef RegisterImpl* Register;
-
-// The implementation of integer registers for z/Architecture.
 
 // z/Architecture registers, see "LINUX for zSeries ELF ABI Supplement", IBM March 2001
 //
@@ -57,6 +52,17 @@
 //   f1,f3,f5,f7 General purpose (volatile)
 //   f8-f15      General purpose (nonvolatile)
 
+
+//===========================
+//===  Integer Registers  ===
+//===========================
+
+// Use Register as shortcut.
+class RegisterImpl;
+typedef RegisterImpl* Register;
+
+// The implementation of integer registers for z/Architecture.
+
 inline Register as_Register(int encoding) {
   return (Register)(long)encoding;
 }
@@ -110,6 +116,11 @@
 CONSTANT_REGISTER_DECLARATION(Register, Z_R14, (14));
 CONSTANT_REGISTER_DECLARATION(Register, Z_R15, (15));
 
+
+//=============================
+//===  Condition Registers  ===
+//=============================
+
 // Use ConditionRegister as shortcut
 class ConditionRegisterImpl;
 typedef ConditionRegisterImpl* ConditionRegister;
@@ -159,7 +170,7 @@
 // dangers of defines.
 // If a particular file has a problem with these defines then it's possible
 // to turn them off in that file by defining
-// DONT_USE_REGISTER_DEFINES. Register_definition_s390.cpp does that
+// DONT_USE_REGISTER_DEFINES. Register_definitions_s390.cpp does that
 // so that it's able to provide real definitions of these registers
 // for use in debuggers and such.
 
@@ -186,6 +197,11 @@
 #define Z_CR ((ConditionRegister)(Z_CR_ConditionRegisterEnumValue))
 #endif // DONT_USE_REGISTER_DEFINES
 
+
+//=========================
+//===  Float Registers  ===
+//=========================
+
 // Use FloatRegister as shortcut
 class FloatRegisterImpl;
 typedef FloatRegisterImpl* FloatRegister;
@@ -263,22 +279,6 @@
 #define Z_F15 ((FloatRegister)(  Z_F15_FloatRegisterEnumValue))
 #endif // DONT_USE_REGISTER_DEFINES
 
-// Need to know the total number of registers of all sorts for SharedInfo.
-// Define a class that exports it.
-
-class ConcreteRegisterImpl : public AbstractRegisterImpl {
- public:
-  enum {
-    number_of_registers =
-      (RegisterImpl::number_of_registers +
-      FloatRegisterImpl::number_of_registers)
-      * 2 // register halves
-      + 1 // condition code register
-  };
-  static const int max_gpr;
-  static const int max_fpr;
-};
-
 // Single, Double and Quad fp reg classes. These exist to map the ADLC
 // encoding for a floating point register, to the FloatRegister number
 // desired by the macroassembler. A FloatRegister is a number between
@@ -329,6 +329,161 @@
 };
 
 
+//==========================
+//===  Vector Registers  ===
+//==========================
+
+// Use VectorRegister as shortcut
+class VectorRegisterImpl;
+typedef VectorRegisterImpl* VectorRegister;
+
+// The implementation of vector registers for z/Architecture.
+
+inline VectorRegister as_VectorRegister(int encoding) {
+  return (VectorRegister)(long)encoding;
+}
+
+class VectorRegisterImpl: public AbstractRegisterImpl {
+ public:
+  enum {
+    number_of_registers     = 32,
+    number_of_arg_registers = 0
+  };
+
+  // construction
+  inline friend VectorRegister as_VectorRegister(int encoding);
+
+  inline VMReg as_VMReg();
+
+  // accessors
+  int encoding() const                                {
+     assert(is_valid(), "invalid register"); return value();
+  }
+
+  bool is_valid() const           { return  0 <= value() && value() < number_of_registers; }
+  bool is_volatile() const        { return true; }
+  bool is_nonvolatile() const     { return false; }
+
+  // Register fields in z/Architecture instructions are 4 bits wide, restricting the
+  // addressable register set size to 16.
+  // The vector register set size is 32, requiring an extension, by one bit, of the
+  // register encoding. This is accomplished by the introduction of a RXB field in the
+  // instruction. RXB = Register eXtension Bits.
+  // The RXB field contains the MSBs (most significant bit) of the vector register numbers
+  // used for this instruction. Assignment of MSB in RBX is by bit position of the
+  // register field in the instruction.
+  // Example:
+  //   The register field starting at bit position 12 in the instruction is assigned RXB bit 0b0100.
+  int64_t RXB_mask(int pos) {
+    if (encoding() >= number_of_registers/2) {
+      switch (pos) {
+        case 8:   return ((int64_t)0b1000) << 8; // actual bit pos: 36
+        case 12:  return ((int64_t)0b0100) << 8; // actual bit pos: 37
+        case 16:  return ((int64_t)0b0010) << 8; // actual bit pos: 38
+        case 32:  return ((int64_t)0b0001) << 8; // actual bit pos: 39
+        default:
+          ShouldNotReachHere();
+      }
+    }
+    return 0;
+  }
+
+  const char* name() const;
+
+  VectorRegister successor() const { return as_VectorRegister(encoding() + 1); }
+};
+
+// The Vector registers of z/Architecture.
+
+CONSTANT_REGISTER_DECLARATION(VectorRegister, vnoreg, (-1));
+
+CONSTANT_REGISTER_DECLARATION(VectorRegister,  Z_V0,  (0));
+CONSTANT_REGISTER_DECLARATION(VectorRegister,  Z_V1,  (1));
+CONSTANT_REGISTER_DECLARATION(VectorRegister,  Z_V2,  (2));
+CONSTANT_REGISTER_DECLARATION(VectorRegister,  Z_V3,  (3));
+CONSTANT_REGISTER_DECLARATION(VectorRegister,  Z_V4,  (4));
+CONSTANT_REGISTER_DECLARATION(VectorRegister,  Z_V5,  (5));
+CONSTANT_REGISTER_DECLARATION(VectorRegister,  Z_V6,  (6));
+CONSTANT_REGISTER_DECLARATION(VectorRegister,  Z_V7,  (7));
+CONSTANT_REGISTER_DECLARATION(VectorRegister,  Z_V8,  (8));
+CONSTANT_REGISTER_DECLARATION(VectorRegister,  Z_V9,  (9));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V10, (10));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V11, (11));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V12, (12));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V13, (13));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V14, (14));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V15, (15));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V16, (16));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V17, (17));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V18, (18));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V19, (19));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V20, (20));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V21, (21));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V22, (22));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V23, (23));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V24, (24));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V25, (25));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V26, (26));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V27, (27));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V28, (28));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V29, (29));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V30, (30));
+CONSTANT_REGISTER_DECLARATION(VectorRegister, Z_V31, (31));
+
+#ifndef DONT_USE_REGISTER_DEFINES
+#define vnoreg ((VectorRegister)(vnoreg_VectorRegisterEnumValue))
+#define Z_V0  ((VectorRegister)(   Z_V0_VectorRegisterEnumValue))
+#define Z_V1  ((VectorRegister)(   Z_V1_VectorRegisterEnumValue))
+#define Z_V2  ((VectorRegister)(   Z_V2_VectorRegisterEnumValue))
+#define Z_V3  ((VectorRegister)(   Z_V3_VectorRegisterEnumValue))
+#define Z_V4  ((VectorRegister)(   Z_V4_VectorRegisterEnumValue))
+#define Z_V5  ((VectorRegister)(   Z_V5_VectorRegisterEnumValue))
+#define Z_V6  ((VectorRegister)(   Z_V6_VectorRegisterEnumValue))
+#define Z_V7  ((VectorRegister)(   Z_V7_VectorRegisterEnumValue))
+#define Z_V8  ((VectorRegister)(   Z_V8_VectorRegisterEnumValue))
+#define Z_V9  ((VectorRegister)(   Z_V9_VectorRegisterEnumValue))
+#define Z_V10 ((VectorRegister)(  Z_V10_VectorRegisterEnumValue))
+#define Z_V11 ((VectorRegister)(  Z_V11_VectorRegisterEnumValue))
+#define Z_V12 ((VectorRegister)(  Z_V12_VectorRegisterEnumValue))
+#define Z_V13 ((VectorRegister)(  Z_V13_VectorRegisterEnumValue))
+#define Z_V14 ((VectorRegister)(  Z_V14_VectorRegisterEnumValue))
+#define Z_V15 ((VectorRegister)(  Z_V15_VectorRegisterEnumValue))
+#define Z_V16 ((VectorRegister)(  Z_V16_VectorRegisterEnumValue))
+#define Z_V17 ((VectorRegister)(  Z_V17_VectorRegisterEnumValue))
+#define Z_V18 ((VectorRegister)(  Z_V18_VectorRegisterEnumValue))
+#define Z_V19 ((VectorRegister)(  Z_V19_VectorRegisterEnumValue))
+#define Z_V20 ((VectorRegister)(  Z_V20_VectorRegisterEnumValue))
+#define Z_V21 ((VectorRegister)(  Z_V21_VectorRegisterEnumValue))
+#define Z_V22 ((VectorRegister)(  Z_V22_VectorRegisterEnumValue))
+#define Z_V23 ((VectorRegister)(  Z_V23_VectorRegisterEnumValue))
+#define Z_V24 ((VectorRegister)(  Z_V24_VectorRegisterEnumValue))
+#define Z_V25 ((VectorRegister)(  Z_V25_VectorRegisterEnumValue))
+#define Z_V26 ((VectorRegister)(  Z_V26_VectorRegisterEnumValue))
+#define Z_V27 ((VectorRegister)(  Z_V27_VectorRegisterEnumValue))
+#define Z_V28 ((VectorRegister)(  Z_V28_VectorRegisterEnumValue))
+#define Z_V29 ((VectorRegister)(  Z_V29_VectorRegisterEnumValue))
+#define Z_V30 ((VectorRegister)(  Z_V30_VectorRegisterEnumValue))
+#define Z_V31 ((VectorRegister)(  Z_V31_VectorRegisterEnumValue))
+#endif // DONT_USE_REGISTER_DEFINES
+
+
+// Need to know the total number of registers of all sorts for SharedInfo.
+// Define a class that exports it.
+
+class ConcreteRegisterImpl : public AbstractRegisterImpl {
+ public:
+  enum {
+    number_of_registers =
+      (RegisterImpl::number_of_registers +
+      FloatRegisterImpl::number_of_registers)
+      * 2 // register halves
+      + 1 // condition code register
+  };
+  static const int max_gpr;
+  static const int max_fpr;
+};
+
+
 // Common register declarations used in assembler code.
 REGISTER_DECLARATION(Register,      Z_EXC_OOP, Z_R2);
 REGISTER_DECLARATION(Register,      Z_EXC_PC,  Z_R3);