jdk8-mips64-public/hotspot: src/cpu/mips/vm/assembler

src/cpu/mips/vm/assembler_mips.cpp@45b68a4063c2 (annotated)

src/cpu/mips/vm/assembler_mips.cpp

Tue, 28 Jun 2016 16:34:47 +0800

author: aoqi<aoqi@loongson.cn>
date: Tue, 28 Jun 2016 16:34:47 +0800
changeset 30: 45b68a4063c2
parent 29: 6c147e7e4605
child 31: f9d3579d1f72
permissions: -rw-r--r--

[Code Reorganization] Moved fast_lock and fast_unlock from mips_64.ad to MacroAssembler.

 /*
  * Copyright (c) 1997, 2014, Oracle and/or its affiliates. All rights reserved.
  * Copyright (c) 2015, 2016, Loongson Technology. 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
  * under the terms of the GNU General Public License version 2 only, as
  * published by the Free Software Foundation.
  *
  * This code is distributed in the hope that it will be useful, but WITHOUT
  * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
  * version 2 for more details (a copy is included in the LICENSE file that
  * accompanied this code).
  *
  * You should have received a copy of the GNU General Public License version
  * 2 along with this work; if not, write to the Free Software Foundation,
  * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
  *
  * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
  * or visit www.oracle.com if you need additional information or have any
  * questions.
  *
  */
 #include "precompiled.hpp"
 #include "asm/assembler.hpp"
 #include "asm/assembler.inline.hpp"
 #include "gc_interface/collectedHeap.inline.hpp"
 #include "interpreter/interpreter.hpp"
 #include "memory/cardTableModRefBS.hpp"
 #include "memory/resourceArea.hpp"
 #include "prims/methodHandles.hpp"
 #include "runtime/biasedLocking.hpp"
 #include "runtime/interfaceSupport.hpp"
 #include "runtime/objectMonitor.hpp"
 #include "runtime/os.hpp"
 #include "runtime/sharedRuntime.hpp"
 #include "runtime/stubRoutines.hpp"
 #ifndef SERIALGC
 #include "gc_implementation/g1/g1CollectedHeap.inline.hpp"
 #include "gc_implementation/g1/g1SATBCardTableModRefBS.hpp"
 #include "gc_implementation/g1/heapRegion.hpp"
 #endif
 #ifdef PRODUCT
 #define BLOCK_COMMENT(str) /* nothing */
 #define STOP(error) stop(error)
 #else
 #define BLOCK_COMMENT(str) block_comment(str)
 #define STOP(error) block_comment(error); stop(error)
 #endif
 #define BIND(label) bind(label); BLOCK_COMMENT(#label ":")
 intptr_t MacroAssembler::i[32] = {0};
 float MacroAssembler::f[32] = {0.0};
 void MacroAssembler::print(outputStream *s) {
 	unsigned int k;
 	for(k=0; k<sizeof(i)/sizeof(i[0]); k++) {
 		s->print_cr("i%d = 0x%.16lx", k, i[k]);
 	}
 	s->cr();
 	for(k=0; k<sizeof(f)/sizeof(f[0]); k++) {
 		s->print_cr("f%d = %f", k, f[k]);
 	}
 	s->cr();
 }
 int MacroAssembler::i_offset(unsigned int k) { return (intptr_t)&((MacroAssembler*)0)->i[k]; }
 int MacroAssembler::f_offset(unsigned int k) { return (intptr_t)&((MacroAssembler*)0)->f[k]; }
 void MacroAssembler::save_registers(MacroAssembler *masm) {
 #define __ masm->
 	for(int k=0; k<32; k++) {
 		__ sw (as_Register(k), A0, i_offset(k));
 	}
 	for(int k=0; k<32; k++) {
 		__ swc1 (as_FloatRegister(k), A0, f_offset(k));
 	}
 #undef __
 }
 void MacroAssembler::restore_registers(MacroAssembler *masm) {
 #define __ masm->
 	for(int k=0; k<32; k++) {
 		__ lw (as_Register(k), A0, i_offset(k));
 	}
 	for(int k=0; k<32; k++) {
 		__ lwc1 (as_FloatRegister(k), A0, f_offset(k));
 	}
 #undef __
 }
 // Implementation of AddressLiteral
 AddressLiteral::AddressLiteral(address target, relocInfo::relocType rtype) {
   _is_lval = false;
   _target = target;
   _rspec = rspec_from_rtype(rtype, target);
 }
 // Implementation of Address
 //FIXME aoqi
 //#ifdef _LP64
 #if 0
 Address Address::make_array(ArrayAddress adr) {
   // Not implementable on 64bit machines
   // Should have been handled higher up the call chain.
   ShouldNotReachHere();
   return Address();
 }
 // exceedingly dangerous constructor
 Address::Address(int disp, address loc, relocInfo::relocType rtype) {
   _base  = noreg;
   _index = noreg;
   _scale = no_scale;
   _disp  = disp;
   switch (rtype) {
     case relocInfo::external_word_type:
       _rspec = external_word_Relocation::spec(loc);
       break;
     case relocInfo::internal_word_type:
       _rspec = internal_word_Relocation::spec(loc);
       break;
     case relocInfo::runtime_call_type:
       // HMM
       _rspec = runtime_call_Relocation::spec();
       break;
     case relocInfo::poll_type:
     case relocInfo::poll_return_type:
       _rspec = Relocation::spec_simple(rtype);
       break;
     case relocInfo::none:
       break;
     default:
       ShouldNotReachHere();
   }
 }
 #else // LP64
 Address Address::make_array(ArrayAddress adr) {
   AddressLiteral base = adr.base();
   Address index = adr.index();
   assert(index._disp == 0, "must not have disp"); // maybe it can?
   Address array(index._base, index._index, index._scale, (intptr_t) base.target());
   array._rspec = base._rspec;
   return array;
 }
 // exceedingly dangerous constructor
 Address::Address(address loc, RelocationHolder spec) {
   _base  = noreg;
   _index = noreg;
   _scale = no_scale;
   _disp  = (intptr_t) loc;
   _rspec = spec;
 }
 #endif // _LP64
 /*
 // Convert the raw encoding form into the form expected by the constructor for
 // Address.  An index of 4 (rsp) corresponds to having no index, so convert
 // that to noreg for the Address constructor.
 Address Address::make_raw(int base, int index, int scale, int disp) {
   bool valid_index = index != rsp->encoding();
   if (valid_index) {
     Address madr(as_Register(base), as_Register(index), (Address::ScaleFactor)scale, in_ByteSize(disp));
     return madr;
   } else {
     Address madr(as_Register(base), noreg, Address::no_scale, in_ByteSize(disp));
     return madr;
   }
 }
 */
 // Implementation of Assembler
 const char *Assembler::ops_name[] = {
 	"special",  "regimm",   "j",      "jal",    "beq",      "bne",      "blez",   "bgtz",
 	"addi",     "addiu",    "slti",   "sltiu",  "andi",     "ori",      "xori",   "lui",
 	"cop0",     "cop1",     "cop2",   "cop3",   "beql",     "bnel",     "bleql",  "bgtzl",
 	"daddi",    "daddiu",   "ldl",    "ldr",    "",         "",         "",       "",
 	"lb",       "lh",       "lwl",    "lw",     "lbu",      "lhu",      "lwr",    "lwu",
 	"sb",       "sh",       "swl",    "sw",     "sdl",      "sdr",      "swr",    "cache",
 	"ll",       "lwc1",     "",       "",       "lld",      "ldc1",     "",       "ld",
 	"sc",       "swc1",     "",       "",       "scd",      "sdc1",     "",       "sd"
 };
 const char* Assembler::special_name[] = {
 	"sll",      "",         "srl",      "sra",      "sllv",     "",         "srlv",     "srav",
 	"jr",       "jalr",     "",         "",         "syscall",  "break",    "",         "sync",
 	"mfhi",     "mthi",     "mflo",     "mtlo",     "dsll",     "",         "dsrl",     "dsra",
 	"mult",     "multu",    "div",      "divu",     "dmult",    "dmultu",   "ddiv",     "ddivu",
 	"add",      "addu",     "sub",      "subu",     "and",      "or",       "xor",      "nor",
 	"",         "",         "slt",      "sltu",     "dadd",     "daddu",    "dsub",     "dsubu",
 	"tge",      "tgeu",     "tlt",      "tltu",     "teq",      "",         "tne",      "",
 	"dsll",     "",         "dsrl",     "dsra",     "dsll32",   "",         "dsrl32",   "dsra32"
 };
 const char* Assembler::regimm_name[] = {
 	"bltz",     "bgez",     "bltzl",    "bgezl",    "",         "",         "",         "",
 	"tgei",     "tgeiu",    "tlti",     "tltiu",    "teqi",     "",         "tnei",     "",
 	"bltzal",   "bgezal",   "bltzall",  "bgezall"
 };
 const char* Assembler::float_name[] = {
 	"add",			"sub",			"mul",			"div",			"sqrt",			"abs",			"mov",			"neg",
 	"round.l",	"trunc.l",	"ceil.l",		"floor.l",	"round.w",  "trunc.w",	"ceil.w",		"floor.w"
 };
 //misleading name, print only branch/jump instruction
 void Assembler::print_instruction(int inst) {
 	const char *s;
 	switch( opcode(inst) ) {
 	default:
 		s = ops_name[opcode(inst)];
 		break;
 	case special_op:
 		s = special_name[special(inst)];
 		break;
 	case regimm_op:
 		s = special_name[rt(inst)];
 		break;
 	}
 	::tty->print("%s", s);
 }
 void MacroAssembler::pd_patch_instruction(address branch, address target) {
   jint& stub_inst = *(jint*) branch;
 /* *
 	move(AT, RA); // dadd
 	emit_long(insn_ORRI(regimm_op, 0, bgezal_op, 1));
 	nop();
         lui(T9, 0); // to be patched
         ori(T9, 0);
 	daddu(T9, T9, RA);
 	move(RA, AT);
 	jr(T9);
  */
   if(special(stub_inst) == dadd_op) {
     jint *pc = (jint *)branch;
     assert(opcode(pc[3]) == lui_op
           && opcode(pc[4]) == ori_op
           && special(pc[5]) == daddu_op, "Not a branch label patch");
     if(!(opcode(pc[3]) == lui_op
           && opcode(pc[4]) == ori_op
           && special(pc[5]) == daddu_op)) { tty->print_cr("Not a branch label patch"); }
     int offset = target - branch;
     if (!is_simm16(offset))
     {
       pc[3] = (pc[3] & 0xffff0000) | high16(offset - 12);
       pc[4] = (pc[4] & 0xffff0000) | low16(offset - 12);
     }
     else
     {
       /* revert to "beq + nop" */
       CodeBuffer cb(branch, 4 * 10);
       MacroAssembler masm(&cb);
 #define __ masm.
       __ b(target);
       __ nop();
       __ nop();
       __ nop();
       __ nop();
       __ nop();
       __ nop();
       __ nop();
     }
     return;
   }
 #ifndef PRODUCT
   if (!is_simm16((target - branch - 4) >> 2))
   {
     tty->print_cr("Illegal patching: target=0x%lx", target);
     int *p = (int *)branch;
     for (int i = -10; i < 10; i++)
     {
        tty->print("0x%lx, ", p[i]);
     }
     tty->print_cr("");
   }
 #endif
   stub_inst = patched_branch(target - branch, stub_inst, 0);
 }
 //without check, maybe fixed
 int Assembler::patched_branch(int dest_pos, int inst, int inst_pos) {
 	int v = (dest_pos - inst_pos - 4)>>2;
 	switch(opcode(inst)) {
 	case j_op:
 	case jal_op:
 		assert(false, "should not use j/jal here");
 		break;
 	default:
 		assert(is_simm16(v), "must be simm16");
 #ifndef PRODUCT
 		if(!is_simm16(v))
 		{
 			tty->print_cr("must be simm16");
 			tty->print_cr("Inst: %lx", inst);
 		}
 #endif
 		v = low16(v);
 		inst &= 0xffff0000;
 		break;
 	}
 	return inst | v;
 }
 int Assembler::branch_destination(int inst, int pos) {
 	int off;
 	switch(opcode(inst)) {
 	case j_op:
 	case jal_op:
 		assert(false, "should not use j/jal here");
 		break;
 	default:
 		off = expand(low16(inst), 15);
 		break;
 	}
 	return off ? pos + 4 + (off<<2) : 0;
 }
 int AbstractAssembler::code_fill_byte() {
 	  return 0x00;                  // illegal instruction 0x00000000
 }
 // Now the Assembler instruction (identical for 32/64 bits)
 void Assembler::lb(Register rt, Address src) {
 	lb(rt, src.base(), src.disp());
 }
 void Assembler::lbu(Register rt, Address src) {
 	lbu(rt, src.base(), src.disp());
 }
 void Assembler::ld(Register rt, Address src){
 	ld(rt, src.base(), src.disp());
 }
 void Assembler::ldl(Register rt, Address src){
 	ldl(rt, src.base(), src.disp());
 }
 void Assembler::ldr(Register rt, Address src){
 	ldr(rt, src.base(), src.disp());
 }
 void Assembler::lh(Register rt, Address src){
 	lh(rt, src.base(), src.disp());
 }
 void Assembler::lhu(Register rt, Address src){
 	lhu(rt, src.base(), src.disp());
 }
 void Assembler::ll(Register rt, Address src){
 	ll(rt, src.base(), src.disp());
 }
 void Assembler::lld(Register rt, Address src){
 	lld(rt, src.base(), src.disp());
 }
 void Assembler::lw(Register rt, Address src){
 	lw(rt, src.base(), src.disp());
 }
 void Assembler::lea(Register rt, Address src) {
 #ifdef _LP64
   daddi(rt, src.base(), src.disp());
 #else
   addi(rt, src.base(), src.disp());
 #endif
 }
 void Assembler::lwl(Register rt, Address src){
 	lwl(rt, src.base(), src.disp());
 }
 void Assembler::lwr(Register rt, Address src){
 	lwr(rt, src.base(), src.disp());
 }
 void Assembler::lwu(Register rt, Address src){
 	lwu(rt, src.base(), src.disp());
 }
 void Assembler::sb(Register rt, Address dst) {
 	sb(rt, dst.base(), dst.disp());
 }
 void Assembler::sc(Register rt, Address dst) {
 	sc(rt, dst.base(), dst.disp());
 }
 void Assembler::scd(Register rt, Address dst) {
 	scd(rt, dst.base(), dst.disp());
 }
 void Assembler::sd(Register rt, Address dst) {
 	sd(rt, dst.base(), dst.disp());
 }
 void Assembler::sdl(Register rt, Address dst) {
 	sdl(rt, dst.base(), dst.disp());
 }
 void Assembler::sdr(Register rt, Address dst) {
 	sdr(rt, dst.base(), dst.disp());
 }
 void Assembler::sh(Register rt, Address dst) {
 	sh(rt, dst.base(), dst.disp());
 }
 void Assembler::sw(Register rt, Address dst) {
 	sw(rt, dst.base(), dst.disp());
 }
 void Assembler::swl(Register rt, Address dst) {
 	swl(rt, dst.base(), dst.disp());
 }
 void Assembler::swr(Register rt, Address dst) {
 	swr(rt, dst.base(), dst.disp());
 }
 void Assembler::lwc1(FloatRegister rt, Address src) {
 	lwc1(rt, src.base(), src.disp());
 }
 void Assembler::ldc1(FloatRegister rt, Address src) {
 	ldc1(rt, src.base(), src.disp());
 }
 void Assembler::swc1(FloatRegister rt, Address dst) {
 	swc1(rt, dst.base(), dst.disp());
 }
 void Assembler::sdc1(FloatRegister rt, Address dst) {
 	sdc1(rt, dst.base(), dst.disp());
 }
 void Assembler::j(address entry) {
 	int dest = ((intptr_t)entry - (((intptr_t)pc() + 4) & 0xf0000000))>>2;
 	emit_long((j_op<<26) | dest);
 	has_delay_slot();
 }
 void Assembler::jal(address entry) {
 	int dest = ((intptr_t)entry - (((intptr_t)pc() + 4) & 0xf0000000))>>2;
 	emit_long((jal_op<<26) | dest);
 	has_delay_slot();
 }
 void MacroAssembler::beq_far(Register rs, Register rt, address entry)
 {
   u_char * cur_pc = pc();
   /* Jin: Near/Far jump */
   if(is_simm16((entry - pc() - 4) / 4))
   {
     Assembler::beq(rs, rt, offset(entry));
   }
   else
   {
     Label not_jump;
     bne(rs, rt, not_jump);
     delayed()->nop();
     b_far(entry);
     delayed()->nop();
     bind(not_jump);
     has_delay_slot();
   }
 }
 void MacroAssembler::beq_far(Register rs, Register rt, Label& L)
 {
   if (L.is_bound()) {
     beq_far(rs, rt, target(L));
   } else {
     u_char * cur_pc = pc();
     Label not_jump;
     bne(rs, rt, not_jump);
     delayed()->nop();
     b_far(L);
     delayed()->nop();
     bind(not_jump);
     has_delay_slot();
   }
 }
 void MacroAssembler::bne_far(Register rs, Register rt, address entry)
 {
   u_char * cur_pc = pc();
   /* Jin: Near/Far jump */
   if(is_simm16((entry - pc() - 4) / 4))
   {
     Assembler::bne(rs, rt, offset(entry));
   }
   else
   {
     Label not_jump;
     beq(rs, rt, not_jump);
     delayed()->nop();
     b_far(entry);
     delayed()->nop();
     bind(not_jump);
     has_delay_slot();
   }
 }
 void MacroAssembler::bne_far(Register rs, Register rt, Label& L)
 {
   if (L.is_bound()) {
     bne_far(rs, rt, target(L));
   } else {
     u_char * cur_pc = pc();
     Label not_jump;
     beq(rs, rt, not_jump);
     delayed()->nop();
     b_far(L);
     delayed()->nop();
     bind(not_jump);
     has_delay_slot();
   }
 }
 void MacroAssembler::b_far(Label& L)
 {
   if (L.is_bound()) {
     b_far(target(L));
   } else {
 	volatile address dest = target(L);
 /*
 MacroAssembler::pd_patch_instruction branch=55651ed514, target=55651ef6d8
 x00000055651ed514: dadd at, ra, zero
 x00000055651ed518: [4110001]bgezal zero, 0x00000055651ed520
 x00000055651ed51c: sll zero, zero, 0
 x00000055651ed520: lui t9, 0x0
 x00000055651ed524: ori t9, t9, 0x21b8
 x00000055651ed528: daddu t9, t9, ra
 x00000055651ed52c: dadd ra, at, zero
 x00000055651ed530: jr t9
 x00000055651ed534: sll zero, zero, 0
 */
 	move(AT, RA);
 	emit_long(insn_ORRI(regimm_op, 0, bgezal_op, 1));
 	nop();
         lui(T9, 0); // to be patched
         ori(T9, T9, 0);
 	daddu(T9, T9, RA);
 	move(RA, AT);
 	jr(T9);
   }
 }
 void MacroAssembler::b_far(address entry)
 {
 	u_char * cur_pc = pc();
 	/* Jin: Near/Far jump */
 	if(is_simm16((entry - pc() - 4) / 4))
 	{
 		b(offset(entry));
 	}
 	else
 	{
 		/* address must be bounded */
 		move(AT, RA);
 	 	emit_long(insn_ORRI(regimm_op, 0, bgezal_op, 1));
 		nop();
 		li32(T9, entry - pc());
 		daddu(T9, T9, RA);
 		move(RA, AT);
 		jr(T9);
 	}
 }
 // Implementation of MacroAssembler
 // First all the versions that have distinct versions depending on 32/64 bit
 // Unless the difference is trivial (1 line or so).
 //#ifndef _LP64
 // 32bit versions
 void MacroAssembler::ld_ptr(Register rt, Register offset, Register base) {
   addu_long(AT, base, offset);
   ld_ptr(rt, 0, AT);
 }
 void MacroAssembler::st_ptr(Register rt, Register offset, Register base) {
   addu_long(AT, base, offset);
   st_ptr(rt, 0, AT);
 }
 void MacroAssembler::ld_long(Register rt, Register offset, Register base) {
   addu_long(AT, base, offset);
   ld_long(rt, 0, AT);
 }
 void MacroAssembler::st_long(Register rt, Register offset, Register base) {
   addu_long(AT, base, offset);
   st_long(rt, 0, AT);
 }
 Address MacroAssembler::as_Address(AddressLiteral adr) {
   return Address(adr.target(), adr.rspec());
 }
 Address MacroAssembler::as_Address(ArrayAddress adr) {
   return Address::make_array(adr);
 }
 // tmp_reg1 and tmp_reg2 should be saved outside of atomic_inc32 (caller saved).
 void MacroAssembler::atomic_inc32(address counter_addr, int inc, Register tmp_reg1, Register tmp_reg2) {
   Label again;
   bind(again);
   sync();
   li(tmp_reg1, counter_addr);
   ll(tmp_reg2, tmp_reg1, 0);
   addi(tmp_reg2, tmp_reg2, inc);
   sc(tmp_reg2, tmp_reg1, 0);
   beq(tmp_reg2, R0, again);
   delayed()->nop();
 }
 int MacroAssembler::biased_locking_enter(Register lock_reg,
                                          Register obj_reg,
                                          Register swap_reg,
                                          Register tmp_reg,
                                          bool swap_reg_contains_mark,
                                          Label& done,
                                          Label* slow_case,
                                          BiasedLockingCounters* counters) {
   assert(UseBiasedLocking, "why call this otherwise?");
   bool need_tmp_reg = false;
   if (tmp_reg == noreg) {
     need_tmp_reg = true;
     tmp_reg = T9;
   }
   assert_different_registers(lock_reg, obj_reg, swap_reg, tmp_reg, AT);
   assert(markOopDesc::age_shift == markOopDesc::lock_bits + markOopDesc::biased_lock_bits, "biased locking makes assumptions about bit layout");
   Address mark_addr      (obj_reg, oopDesc::mark_offset_in_bytes());
   Address saved_mark_addr(lock_reg, 0);
   // Biased locking
   // See whether the lock is currently biased toward our thread and
   // whether the epoch is still valid
   // Note that the runtime guarantees sufficient alignment of JavaThread
   // pointers to allow age to be placed into low bits
   // First check to see whether biasing is even enabled for this object
   Label cas_label;
   int null_check_offset = -1;
   if (!swap_reg_contains_mark) {
     null_check_offset = offset();
     ld_ptr(swap_reg, mark_addr);
   }
   if (need_tmp_reg) {
     push(tmp_reg);
   }
   move(tmp_reg, swap_reg);
   andi(tmp_reg, tmp_reg, markOopDesc::biased_lock_mask_in_place);
 #ifdef _LP64
   daddi(AT, R0,markOopDesc::biased_lock_pattern);
   dsub(AT, AT, tmp_reg);
 #else
   addi(AT, R0,markOopDesc::biased_lock_pattern);
   sub(AT, AT, tmp_reg);
 #endif
   if (need_tmp_reg) {
     // popl(tmp_reg);
     pop(tmp_reg);
   }
   //jcc(Assembler::notEqual, cas_label);
   bne(AT, R0, cas_label);
   delayed()->nop();
   // The bias pattern is present in the object's header. Need to check
   // whether the bias owner and the epoch are both still current.
   // Note that because there is no current thread register on x86 we
   // need to store off the mark word we read out of the object to
   // avoid reloading it and needing to recheck invariants below. This
   // store is unfortunate but it makes the overall code shorter and
   // simpler.
   // movl(saved_mark_addr, swap_reg);
   st_ptr(swap_reg, saved_mark_addr);
   if (need_tmp_reg) {
     push(tmp_reg);
   }
   get_thread(tmp_reg);
   //xorl(swap_reg, tmp_reg);
   xorr(swap_reg,swap_reg, tmp_reg);
   if (swap_reg_contains_mark) {
     null_check_offset = offset();
   }
   //add for compressedoops
   //printf("load_prototype l727\n");
   load_prototype_header(tmp_reg, obj_reg);
   xorr(swap_reg, swap_reg, tmp_reg);
   // andl(swap_reg, ~((int) markOopDesc::age_mask_in_place));
   move(AT, ~((int) markOopDesc::age_mask_in_place));
   andr(swap_reg,swap_reg,AT);
   if (PrintBiasedLockingStatistics) {
     Label L;
     bne(swap_reg, R0, L);
     delayed()->nop();
     atomic_inc32((address)BiasedLocking::biased_lock_entry_count_addr(), 1, AT, tmp_reg);
     bind(L);
   }
   if (need_tmp_reg) {
     //popl(tmp_reg);
     pop(tmp_reg);
   }
   //  jcc(Assembler::equal, done);
   //FIXME, equal is for what ,there is no cmp or test here? @jerome
   //beq(tmp_reg,R0, done);
   beq(swap_reg,R0, done);
   delayed()->nop();
   Label try_revoke_bias;
   Label try_rebias;
   // At this point we know that the header has the bias pattern and
   // that we are not the bias owner in the current epoch. We need to
   // figure out more details about the state of the header in order to
   // know what operations can be legally performed on the object's
   // header.
   // If the low three bits in the xor result aren't clear, that means
   // the prototype header is no longer biased and we have to revoke
   // the bias on this object.
   //testl(swap_reg, markOopDesc::biased_lock_mask_in_place);
   //jcc(Assembler::notZero, try_revoke_bias);
   move(AT, markOopDesc::biased_lock_mask_in_place);
   andr(AT,swap_reg,AT );
   bne(AT,R0,try_revoke_bias);
   delayed()->nop();
   // Biasing is still enabled for this data type. See whether the
   // epoch of the current bias is still valid, meaning that the epoch
   // bits of the mark word are equal to the epoch bits of the
   // prototype header. (Note that the prototype header's epoch bits
   // only change at a safepoint.) If not, attempt to rebias the object
   // toward the current thread. Note that we must be absolutely sure
   // that the current epoch is invalid in order to do this because
   // otherwise the manipulations it performs on the mark word are
   // illegal.
   // testl(swap_reg, markOopDesc::epoch_mask_in_place);
   //jcc(Assembler::notZero, try_rebias);
   move(AT, markOopDesc::epoch_mask_in_place);
   andr(AT,swap_reg,AT);
   bne(AT,R0,try_rebias);
   delayed()->nop();
   // The epoch of the current bias is still valid but we know nothing
   // about the owner; it might be set or it might be clear. Try to
   // acquire the bias of the object using an atomic operation. If this
   // fails we will go in to the runtime to revoke the object's bias.
   // Note that we first construct the presumed unbiased header so we
   // don't accidentally blow away another thread's valid bias.
   ld_ptr(swap_reg, saved_mark_addr);
   //  andl(swap_reg,markOopDesc::biased_lock_mask_in_place | markOopDesc::age_mask_in_place | markOopDesc::epoch_mask_in_place);
   move(AT, markOopDesc::biased_lock_mask_in_place | markOopDesc::age_mask_in_place | markOopDesc::epoch_mask_in_place);
   andr(swap_reg,swap_reg,AT);
   if (need_tmp_reg) {
     // pushl(tmp_reg);
     push(tmp_reg);
   }
   get_thread(tmp_reg);
   //orl(tmp_reg, swap_reg);
   orr(tmp_reg,tmp_reg, swap_reg);
   //if (os::is_MP()) {
   // lock();
   //}
   //cmpxchg(tmp_reg, Address(obj_reg));
   // what is store in eax now ? @jerome,see the entry of the func, swap_reg!
   cmpxchg(tmp_reg, Address(obj_reg, 0),swap_reg);
   if (need_tmp_reg) {
     //popl(tmp_reg);
     pop(tmp_reg);
   }
   // If the biasing toward our thread failed, this means that
   // another thread succeeded in biasing it toward itself and we
   // need to revoke that bias. The revocation will occur in the
   // interpreter runtime in the slow case.
   if (PrintBiasedLockingStatistics) {
     Label L;
     bne(AT, R0, L);
     delayed()->nop();
     push(tmp_reg);
     push(A0);
     atomic_inc32((address)BiasedLocking::anonymously_biased_lock_entry_count_addr(), 1, A0, tmp_reg);
     pop(A0);
     pop(tmp_reg);
     bind(L);
   }
   if (slow_case != NULL) {
     //jcc(Assembler::notZero, *slow_case);
     beq_far(AT,R0, *slow_case);
     delayed()->nop();
   }
   //jmp(done);
   b(done);
   delayed()->nop();
   bind(try_rebias);
   // At this point we know the epoch has expired, meaning that the
   // current "bias owner", if any, is actually invalid. Under these
   // circumstances _only_, we are allowed to use the current header's
   // value as the comparison value when doing the cas to acquire the
   // bias in the current epoch. In other words, we allow transfer of
   // the bias from one thread to another directly in this situation.
   //
   // FIXME: due to a lack of registers we currently blow away the age
   // bits in this situation. Should attempt to preserve them.
   if (need_tmp_reg) {
     // pushl(tmp_reg);
     push(tmp_reg);
   }
   get_thread(tmp_reg);
   //add for compressedoops
   load_prototype_header(swap_reg, obj_reg);
   orr(tmp_reg,tmp_reg,swap_reg);
   ld_ptr(swap_reg, saved_mark_addr);
   // if (os::is_MP()) {
   //  lock();
   //}
   // cmpxchg(tmp_reg, Address(obj_reg));
   cmpxchg(tmp_reg, Address(obj_reg, 0),swap_reg);
   if (need_tmp_reg) {
     //    popl(tmp_reg);
     pop(tmp_reg);
   }
   // If the biasing toward our thread failed, then another thread
   // succeeded in biasing it toward itself and we need to revoke that
   // bias. The revocation will occur in the runtime in the slow case.
   if (PrintBiasedLockingStatistics) {
     Label L;
     bne(AT, R0, L);
     delayed()->nop();
     push(AT);
     push(tmp_reg);
     atomic_inc32((address)BiasedLocking::rebiased_lock_entry_count_addr(), 1, AT, tmp_reg);
     pop(tmp_reg);
     pop(AT);
     bind(L);
   }
   if (slow_case != NULL) {
     //jcc(Assembler::notZero, *slow_case);
     beq_far(AT,R0, *slow_case);
     delayed()->nop();
   }
   //jmp(done);
   b(done);
   delayed()->nop();
   bind(try_revoke_bias);
   // The prototype mark in the klass doesn't have the bias bit set any
   // more, indicating that objects of this data type are not supposed
   // to be biased any more. We are going to try to reset the mark of
   // this object to the prototype value and fall through to the
   // CAS-based locking scheme. Note that if our CAS fails, it means
   // that another thread raced us for the privilege of revoking the
   // bias of this particular object, so it's okay to continue in the
   // normal locking code.
   //
   // FIXME: due to a lack of registers we currently blow away the age
   // bits in this situation. Should attempt to preserve them.
   // movl(swap_reg, saved_mark_addr);
   ld_ptr(swap_reg, saved_mark_addr);
   if (need_tmp_reg) {
     //pushl(tmp_reg);
     push(tmp_reg);
   }
   //add for compressedoops
   load_prototype_header(tmp_reg, obj_reg);
   //if (os::is_MP()) {
   // lock();
   //}
   //cmpxchg(tmp_reg, Address(obj_reg));
   cmpxchg(tmp_reg, Address(obj_reg, 0),swap_reg);
   if (need_tmp_reg) {
     //popl(tmp_reg);
     pop(tmp_reg);
   }
   // Fall through to the normal CAS-based lock, because no matter what
   // the result of the above CAS, some thread must have succeeded in
   // removing the bias bit from the object's header.
   if (PrintBiasedLockingStatistics) {
     Label L;
     bne(AT, R0, L);
     delayed()->nop();
     push(AT);
     push(tmp_reg);
     atomic_inc32((address)BiasedLocking::revoked_lock_entry_count_addr(), 1, AT, tmp_reg);
     pop(tmp_reg);
     pop(AT);
     bind(L);
   }
   bind(cas_label);
   return null_check_offset;
 }
 void MacroAssembler::biased_locking_exit(Register obj_reg, Register temp_reg, Label& done) {
   assert(UseBiasedLocking, "why call this otherwise?");
   // Check for biased locking unlock case, which is a no-op
   // Note: we do not have to check the thread ID for two reasons.
   // First, the interpreter checks for IllegalMonitorStateException at
   // a higher level. Second, if the bias was revoked while we held the
   // lock, the object could not be rebiased toward another thread, so
   // the bias bit would be clear.
 #ifdef _LP64
   ld(temp_reg, Address(obj_reg, oopDesc::mark_offset_in_bytes()));
   andi(temp_reg, temp_reg, markOopDesc::biased_lock_mask_in_place);
   daddi(AT, R0, markOopDesc::biased_lock_pattern);
 #else
   lw(temp_reg, Address(obj_reg, oopDesc::mark_offset_in_bytes()));
   andi(temp_reg, temp_reg, markOopDesc::biased_lock_mask_in_place);
   addi(AT, R0, markOopDesc::biased_lock_pattern);
 #endif
   beq(AT, temp_reg, done);
   delayed()->nop();
 }
 // NOTE: we dont increment the SP after call like the x86 version, maybe this is a problem, FIXME.
 // by yjl 6/27/2005
 // the stack pointer adjustment is needed. see InterpreterMacroAssembler::super_call_VM_leaf
 // by yjl 7/11/2005
 // this method will handle the stack problem, you need not to preserve the stack space for the argument now
 // by yjl 8/1/2005
 void MacroAssembler::call_VM_leaf_base(address entry_point,
     int number_of_arguments) {
   //call(RuntimeAddress(entry_point));
   //increment(rsp, number_of_arguments * wordSize);
   Label L, E;
   assert(number_of_arguments <= 4, "just check");
   andi(AT, SP, 0xf);
   beq(AT, R0, L);
   delayed()->nop();
   daddi(SP, SP, -8);
   {
 	call(entry_point, relocInfo::runtime_call_type);
 	delayed()->nop();
   }
   daddi(SP, SP, 8);
   b(E);
   delayed()->nop();
   bind(L);
   {
 	call(entry_point, relocInfo::runtime_call_type);
 	delayed()->nop();
   }
   bind(E);
 }
 void MacroAssembler::jmp(address entry) {
   li48(T9, (long)entry);
   jr(T9);
 }
 void MacroAssembler::jmp(address entry, relocInfo::relocType rtype) {
   switch (rtype) {
     case relocInfo::runtime_call_type:
     case relocInfo::none:
       jmp(entry);
       break;
     default:
       {
 	InstructionMark im(this);
 	relocate(rtype);
 	li48(T9, (long)entry);
 	jr(T9);
       }
       break;
   }
 }
 void MacroAssembler::call(address entry) {
 // c/c++ code assume T9 is entry point, so we just always move entry to t9
 // maybe there is some more graceful method to handle this. FIXME
 // by yjl 6/27/2005
 // For more info, see class NativeCall.
 #ifndef _LP64
   move(T9, (int)entry);
 #else
   li48(T9, (long)entry);
 #endif
   jalr(T9);
 }
 void MacroAssembler::call(address entry, relocInfo::relocType rtype) {
   switch (rtype) {
     case relocInfo::runtime_call_type:
     case relocInfo::none:
       call(entry);
       break;
     default:
       {
 	InstructionMark im(this);
 	relocate(rtype);
 	call(entry);
       }
       break;
   }
 }
 void MacroAssembler::call(address entry, RelocationHolder& rh)
 {
   switch (rh.type()) {
     case relocInfo::runtime_call_type:
     case relocInfo::none:
       call(entry);
       break;
     default:
       {
 	InstructionMark im(this);
 	relocate(rh);
 	call(entry);
       }
       break;
   }
 }
 void MacroAssembler::ic_call(address entry) {
 	RelocationHolder rh = virtual_call_Relocation::spec(pc());
 	li64(IC_Klass, (long)Universe::non_oop_word());
 	assert(entry != NULL, "call most probably wrong");
 	InstructionMark im(this);
 	relocate(rh);
 	li48(T9, (long)entry);
 	jalr(T9);
 	delayed()->nop();
 }
 void MacroAssembler::c2bool(Register r) {
   Label L;
   Assembler::beq(r, R0, L);
   delayed()->nop();
   move(r, 1);
   bind(L);
 }
 #ifndef PRODUCT
 extern "C" void findpc(intptr_t x);
 #endif
 void MacroAssembler::debug32(int rdi, int rsi, int rbp, int rsp, int rbx, int rdx, int rcx, int rax, int eip, char* msg) {
   // In order to get locks to work, we need to fake a in_VM state
   JavaThread* thread = JavaThread::current();
   JavaThreadState saved_state = thread->thread_state();
   thread->set_thread_state(_thread_in_vm);
   if (ShowMessageBoxOnError) {
     JavaThread* thread = JavaThread::current();
     JavaThreadState saved_state = thread->thread_state();
     thread->set_thread_state(_thread_in_vm);
     if (CountBytecodes || TraceBytecodes || StopInterpreterAt) {
       ttyLocker ttyl;
       BytecodeCounter::print();
     }
     // To see where a verify_oop failed, get $ebx+40/X for this frame.
     // This is the value of eip which points to where verify_oop will return.
     if (os::message_box(msg, "Execution stopped, print registers?")) {
       ttyLocker ttyl;
       tty->print_cr("eip = 0x%08x", eip);
 #ifndef PRODUCT
       tty->cr();
       findpc(eip);
       tty->cr();
 #endif
       tty->print_cr("rax, = 0x%08x", rax);
       tty->print_cr("rbx, = 0x%08x", rbx);
       tty->print_cr("rcx = 0x%08x", rcx);
       tty->print_cr("rdx = 0x%08x", rdx);
       tty->print_cr("rdi = 0x%08x", rdi);
       tty->print_cr("rsi = 0x%08x", rsi);
       tty->print_cr("rbp, = 0x%08x", rbp);
       tty->print_cr("rsp = 0x%08x", rsp);
       BREAKPOINT;
     }
   } else {
     ttyLocker ttyl;
     ::tty->print_cr("=============== DEBUG MESSAGE: %s ================\n", msg);
     assert(false, "DEBUG MESSAGE");
   }
   ThreadStateTransition::transition(thread, _thread_in_vm, saved_state);
 }
 void MacroAssembler::debug(char* msg/*, RegistersForDebugging* regs*/) {
   if ( ShowMessageBoxOnError ) {
     JavaThreadState saved_state = JavaThread::current()->thread_state();
     JavaThread::current()->set_thread_state(_thread_in_vm);
     {
       // In order to get locks work, we need to fake a in_VM state
       ttyLocker ttyl;
       ::tty->print_cr("EXECUTION STOPPED: %s\n", msg);
       if (CountBytecodes || TraceBytecodes || StopInterpreterAt) {
 	BytecodeCounter::print();
       }
       //			if (os::message_box(msg, "Execution stopped, print registers?"))
       //				regs->print(::tty);
     }
     ThreadStateTransition::transition(JavaThread::current(), _thread_in_vm, saved_state);
   }
   else
     ::tty->print_cr("=============== DEBUG MESSAGE: %s ================\n", msg);
 }
 void MacroAssembler::stop(const char* msg) {
   li(A0, (long)msg);
 #ifndef _LP64
   //reserver space for argument. added by yjl 7/10/2005
   addiu(SP, SP, - 1 * wordSize);
 #endif
   call(CAST_FROM_FN_PTR(address, MacroAssembler::debug), relocInfo::runtime_call_type);
   delayed()->nop();
 #ifndef _LP64
   //restore space for argument
   addiu(SP, SP, 1 * wordSize);
 #endif
   brk(17);
 }
 void MacroAssembler::warn(const char* msg) {
 #ifdef _LP64
   pushad();
   li(A0, (long)msg);
   call(CAST_FROM_FN_PTR(address, MacroAssembler::debug), relocInfo::runtime_call_type);
   delayed()->nop();
   popad();
 #else
   pushad();
   addi(SP, SP, -4);
   sw(A0, SP, -1 * wordSize);
   li(A0, (long)msg);
   addi(SP, SP, -1 * wordSize);
   call(CAST_FROM_FN_PTR(address, MacroAssembler::debug), relocInfo::runtime_call_type);
   delayed()->nop();
   addi(SP, SP, 1 * wordSize);
   lw(A0, SP, -1 * wordSize);
   addi(SP, SP, 4);
   popad();
 #endif
 }
 void MacroAssembler::print_reg(Register reg) {
 /*
 char *s = getenv("PRINT_REG");
 if (s == NULL)
   return;
 if (strcmp(s, "1") != 0)
   return;
 */
   void * cur_pc = pc();
   pushad();
   NOT_LP64(push(FP);)
   li(A0, (long)reg->name());
   if (reg == SP)
     addiu(A1, SP, wordSize * 23); //23 registers saved in pushad()
   else if (reg == A0)
     ld(A1, SP, wordSize * 19); //A0 has been modified by li(A0, (long)reg->name()). Ugly Code!
   else
     move(A1, reg);
   li(A2, (long)cur_pc);
   call(CAST_FROM_FN_PTR(address, SharedRuntime::print_reg_with_pc),relocInfo::runtime_call_type);
   delayed()->nop();
   NOT_LP64(pop(FP);)
   popad();
 /*
   pushad();
 #ifdef _LP64
   if (reg == SP)
     addiu(A0, SP, wordSize * 23); //23 registers saved in pushad()
   else
     move(A0, reg);
   call(CAST_FROM_FN_PTR(address, SharedRuntime::print_long),relocInfo::runtime_call_type);
   delayed()->nop();
 #else
   push(FP);
   move(A0, reg);
   dsrl32(A1, reg, 0);
   //call(CAST_FROM_FN_PTR(address, SharedRuntime::print_int),relocInfo::runtime_call_type);
   call(CAST_FROM_FN_PTR(address, SharedRuntime::print_long),relocInfo::runtime_call_type);
   delayed()->nop();
   pop(FP);
 #endif
   popad();
   pushad();
   NOT_LP64(push(FP);)
   char b[50];
   sprintf((char *)b, " pc: %p\n",cur_pc);
   li(A0, (long)(char *)b);
   call(CAST_FROM_FN_PTR(address, SharedRuntime::print_str),relocInfo::runtime_call_type);
   delayed()->nop();
   NOT_LP64(pop(FP);)
   popad();
 */
 }
 void MacroAssembler::print_reg(FloatRegister reg) {
   void * cur_pc = pc();
   pushad();
   NOT_LP64(push(FP);)
   li(A0, (long)reg->name());
   call(CAST_FROM_FN_PTR(address, SharedRuntime::print_str),relocInfo::runtime_call_type);
   delayed()->nop();
   NOT_LP64(pop(FP);)
   popad();
   pushad();
   NOT_LP64(push(FP);)
 #if 1
   move(FP, SP);
   move(AT, -(StackAlignmentInBytes));
   andr(SP , SP , AT);
   mov_d(F12, reg);
   call(CAST_FROM_FN_PTR(address, SharedRuntime::print_double),relocInfo::runtime_call_type);
   delayed()->nop();
   move(SP, FP);
 #else
   mov_s(F12, reg);
   //call(CAST_FROM_FN_PTR(address, SharedRuntime::print_float),relocInfo::runtime_call_type);
   //delayed()->nop();
 #endif
   NOT_LP64(pop(FP);)
   popad();
 #if 0
   pushad();
   NOT_LP64(push(FP);)
   char* b = new char[50];
   sprintf(b, " pc: %p\n", cur_pc);
   li(A0, (long)b);
   call(CAST_FROM_FN_PTR(address, SharedRuntime::print_str),relocInfo::runtime_call_type);
   delayed()->nop();
   NOT_LP64(pop(FP);)
   popad();
 #endif
 }
 void MacroAssembler::increment(Register reg, int imm) {
   if (!imm) return;
   if (is_simm16(imm)) {
 #ifdef _LP64
     daddiu(reg, reg, imm);
 #else
     addiu(reg, reg, imm);
 #endif
   } else {
     move(AT, imm);
 #ifdef _LP64
     daddu(reg, reg, AT);
 #else
     addu(reg, reg, AT);
 #endif
   }
 }
 void MacroAssembler::decrement(Register reg, int imm) {
 	increment(reg, -imm);
 }
 void MacroAssembler::call_VM(Register oop_result,
                              address entry_point,
                              bool check_exceptions) {
   call_VM_helper(oop_result, entry_point, 0, check_exceptions);
 }
 void MacroAssembler::call_VM(Register oop_result,
                              address entry_point,
                              Register arg_1,
                              bool check_exceptions) {
   if (arg_1!=A1) move(A1, arg_1);
   call_VM_helper(oop_result, entry_point, 1, check_exceptions);
 }
 void MacroAssembler::call_VM(Register oop_result,
                              address entry_point,
                              Register arg_1,
                              Register arg_2,
                              bool check_exceptions) {
   if (arg_1!=A1) move(A1, arg_1);
   if (arg_2!=A2) move(A2, arg_2);
   assert(arg_2 != A1, "smashed argument");
   call_VM_helper(oop_result, entry_point, 2, check_exceptions);
 }
 void MacroAssembler::call_VM(Register oop_result,
                              address entry_point,
                              Register arg_1,
                              Register arg_2,
                              Register arg_3,
                              bool check_exceptions) {
   if (arg_1!=A1) move(A1, arg_1);
   if (arg_2!=A2) move(A2, arg_2); assert(arg_2 != A1, "smashed argument");
   if (arg_3!=A3) move(A3, arg_3); assert(arg_3 != A1 && arg_3 != A2, "smashed argument");
   call_VM_helper(oop_result, entry_point, 3, check_exceptions);
 }
 void MacroAssembler::call_VM(Register oop_result,
                              Register last_java_sp,
                              address entry_point,
                              int number_of_arguments,
                              bool check_exceptions) {
   call_VM_base(oop_result, NOREG, last_java_sp, entry_point, number_of_arguments, check_exceptions);
 }
 void MacroAssembler::call_VM(Register oop_result,
                              Register last_java_sp,
                              address entry_point,
                              Register arg_1,
                              bool check_exceptions) {
   if (arg_1 != A1) move(A1, arg_1);
   call_VM(oop_result, last_java_sp, entry_point, 1, check_exceptions);
 }
 void MacroAssembler::call_VM(Register oop_result,
                              Register last_java_sp,
                              address entry_point,
                              Register arg_1,
                              Register arg_2,
                              bool check_exceptions) {
   if (arg_1 != A1) move(A1, arg_1);
   if (arg_2 != A2) move(A2, arg_2); assert(arg_2 != A1, "smashed argument");
   call_VM(oop_result, last_java_sp, entry_point, 2, check_exceptions);
 }
 void MacroAssembler::call_VM(Register oop_result,
                              Register last_java_sp,
                              address entry_point,
                              Register arg_1,
                              Register arg_2,
                              Register arg_3,
                              bool check_exceptions) {
   if (arg_1 != A1) move(A1, arg_1);
   if (arg_2 != A2) move(A2, arg_2); assert(arg_2 != A1, "smashed argument");
   if (arg_3 != A3) move(A3, arg_3); assert(arg_3 != A1 && arg_3 != A2, "smashed argument");
   call_VM(oop_result, last_java_sp, entry_point, 3, check_exceptions);
 }
 void MacroAssembler::call_VM_base(Register oop_result,
                                   Register java_thread,
                                   Register last_java_sp,
                                   address  entry_point,
                                   int      number_of_arguments,
 				  bool     check_exceptions) {
   address before_call_pc;
   // determine java_thread register
   if (!java_thread->is_valid()) {
 #ifndef OPT_THREAD
     java_thread = T2;
     get_thread(java_thread);
 #else
     java_thread = TREG;
 #endif
   }
   // determine last_java_sp register
   if (!last_java_sp->is_valid()) {
     last_java_sp = SP;
   }
   // debugging support
   assert(number_of_arguments >= 0   , "cannot have negative number of arguments");
   assert(number_of_arguments <= 4   , "cannot have negative number of arguments");
   assert(java_thread != oop_result  , "cannot use the same register for java_thread & oop_result");
   assert(java_thread != last_java_sp, "cannot use the same register for java_thread & last_java_sp");
   assert(last_java_sp != FP, "this code doesn't work for last_java_sp == fp, which currently can't portably work anyway since C2 doesn't save ebp");
   // set last Java frame before call
   before_call_pc = (address)pc();
   set_last_Java_frame(java_thread, last_java_sp, FP, before_call_pc);
   // do the call
   move(A0, java_thread);
   call(entry_point, relocInfo::runtime_call_type);
   delayed()->nop();
   // restore the thread (cannot use the pushed argument since arguments
   // may be overwritten by C code generated by an optimizing compiler);
   // however can use the register value directly if it is callee saved.
 #ifndef OPT_THREAD
   if (java_thread >=S0 && java_thread <=S7) {
 #ifdef ASSERT
     { Label L;
       get_thread(AT);
       beq(java_thread, AT, L);
       delayed()->nop();
       stop("MacroAssembler::call_VM_base: edi not callee saved?");
       bind(L);
     }
 #endif
   } else {
     get_thread(java_thread);
   }
 #endif
   // discard thread and arguments
   ld_ptr(SP, java_thread, in_bytes(JavaThread::last_Java_sp_offset()));
   // reset last Java frame
   reset_last_Java_frame(java_thread, false, true);
   check_and_handle_popframe(java_thread);
   check_and_handle_earlyret(java_thread);
   if (check_exceptions) {
     // check for pending exceptions (java_thread is set upon return)
     Label L;
 #ifdef _LP64
     ld(AT, java_thread, in_bytes(Thread::pending_exception_offset()));
 #else
     lw(AT, java_thread, in_bytes(Thread::pending_exception_offset()));
 #endif
     beq(AT, R0, L);
     delayed()->nop();
     li(AT, before_call_pc);
     push(AT);
     jmp(StubRoutines::forward_exception_entry(), relocInfo::runtime_call_type);
     delayed()->nop();
     bind(L);
   }
   // get oop result if there is one and reset the value in the thread
   if (oop_result->is_valid()) {
 #ifdef _LP64
     ld(oop_result, java_thread, in_bytes(JavaThread::vm_result_offset()));
     sd(R0, java_thread, in_bytes(JavaThread::vm_result_offset()));
 #else
     lw(oop_result, java_thread, in_bytes(JavaThread::vm_result_offset()));
     sw(R0, java_thread, in_bytes(JavaThread::vm_result_offset()));
 #endif
     verify_oop(oop_result);
   }
 }
 void MacroAssembler::call_VM_helper(Register oop_result, address entry_point, int number_of_arguments, bool check_exceptions) {
   move(V0, SP);
   //we also reserve space for java_thread here
 #ifndef _LP64
   daddi(SP, SP, (1 + number_of_arguments) * (- wordSize));
 #endif
   move(AT, -(StackAlignmentInBytes));
   andr(SP, SP, AT);
   call_VM_base(oop_result, NOREG, V0, entry_point, number_of_arguments, check_exceptions);
 }
 void MacroAssembler::call_VM_leaf(address entry_point, int number_of_arguments) {
   call_VM_leaf_base(entry_point, number_of_arguments);
 }
 void MacroAssembler::call_VM_leaf(address entry_point, Register arg_0) {
   if (arg_0 != A0) move(A0, arg_0);
   call_VM_leaf(entry_point, 1);
 }
 void MacroAssembler::call_VM_leaf(address entry_point, Register arg_0, Register arg_1) {
   if (arg_0 != A0) move(A0, arg_0);
   if (arg_1 != A1) move(A1, arg_1); assert(arg_1 != A0, "smashed argument");
   call_VM_leaf(entry_point, 2);
 }
 void MacroAssembler::call_VM_leaf(address entry_point, Register arg_0, Register arg_1, Register arg_2) {
   if (arg_0 != A0) move(A0, arg_0);
   if (arg_1 != A1) move(A1, arg_1); assert(arg_1 != A0, "smashed argument");
   if (arg_2 != A2) move(A2, arg_2); assert(arg_2 != A0 && arg_2 != A1, "smashed argument");
   call_VM_leaf(entry_point, 3);
 }
 void MacroAssembler::super_call_VM_leaf(address entry_point) {
 	MacroAssembler::call_VM_leaf_base(entry_point, 0);
 }
 void MacroAssembler::super_call_VM_leaf(address entry_point,
                                                    Register arg_1) {
   if (arg_1 != A0) move(A0, arg_1);
   MacroAssembler::call_VM_leaf_base(entry_point, 1);
 }
 void MacroAssembler::super_call_VM_leaf(address entry_point,
                                                    Register arg_1,
                                                    Register arg_2) {
   if (arg_1 != A0) move(A0, arg_1);
   if (arg_2 != A1) move(A1, arg_2); assert(arg_2 != A0, "smashed argument");
   MacroAssembler::call_VM_leaf_base(entry_point, 2);
 }
 void MacroAssembler::super_call_VM_leaf(address entry_point,
                                                    Register arg_1,
                                                    Register arg_2,
                                                    Register arg_3) {
   if (arg_1 != A0) move(A0, arg_1);
   if (arg_2 != A1) move(A1, arg_2); assert(arg_2 != A0, "smashed argument");
   if (arg_3 != A2) move(A2, arg_3); assert(arg_3 != A0 && arg_3 != A1, "smashed argument");
   MacroAssembler::call_VM_leaf_base(entry_point, 3);
 }
 void MacroAssembler::check_and_handle_earlyret(Register java_thread) {
 }
 void MacroAssembler::check_and_handle_popframe(Register java_thread) {
 }
 void MacroAssembler::null_check(Register reg, int offset) {
   if (needs_explicit_null_check(offset)) {
     // provoke OS NULL exception if reg = NULL by
     // accessing M[reg] w/o changing any (non-CC) registers
     // NOTE: cmpl is plenty here to provoke a segv
     lw(AT, reg, 0);
 /* Jin
     nop();
     nop();
     nop();
 */
     // Note: should probably use testl(rax, Address(reg, 0));
     //       may be shorter code (however, this version of
     //       testl needs to be implemented first)
   } else {
     // nothing to do, (later) access of M[reg + offset]
     // will provoke OS NULL exception if reg = NULL
   }
 }
 void MacroAssembler::enter() {
   push2(RA, FP);
   move(FP, SP);
 }
 void MacroAssembler::leave() {
 #ifndef _LP64
   //move(SP, FP);
   //pop2(FP, RA);
   addi(SP, FP, 2 * wordSize);
   lw(RA, SP, - 1 * wordSize);
   lw(FP, SP, - 2 * wordSize);
 #else
   daddi(SP, FP, 2 * wordSize);
   ld(RA, SP, - 1 * wordSize);
   ld(FP, SP, - 2 * wordSize);
 #endif
 }
 /*
 void MacroAssembler::os_breakpoint() {
   // instead of directly emitting a breakpoint, call os:breakpoint for better debugability
   // (e.g., MSVC can't call ps() otherwise)
   call(RuntimeAddress(CAST_FROM_FN_PTR(address, os::breakpoint)));
 }
 */
 void MacroAssembler::reset_last_Java_frame(Register java_thread, bool clear_fp, bool clear_pc) {
   // determine java_thread register
   if (!java_thread->is_valid()) {
 #ifndef OPT_THREAD
     java_thread = T1;
     get_thread(java_thread);
 #else
     java_thread = TREG;
 #endif
   }
   // we must set sp to zero to clear frame
   st_ptr(R0, java_thread, in_bytes(JavaThread::last_Java_sp_offset()));
   // must clear fp, so that compiled frames are not confused; it is possible
   // that we need it only for debugging
   if(clear_fp)
     st_ptr(R0, java_thread, in_bytes(JavaThread::last_Java_fp_offset()));
   if (clear_pc)
     st_ptr(R0, java_thread, in_bytes(JavaThread::last_Java_pc_offset()));
 }
 void MacroAssembler::reset_last_Java_frame(bool clear_fp,
                                            bool clear_pc) {
   Register thread = TREG;
 #ifndef OPT_THREAD
   get_thread(thread);
 #endif
   // we must set sp to zero to clear frame
   sd(R0, Address(thread, JavaThread::last_Java_sp_offset()));
   // must clear fp, so that compiled frames are not confused; it is
   // possible that we need it only for debugging
   if (clear_fp) {
     sd(R0, Address(thread, JavaThread::last_Java_fp_offset()));
   }
   if (clear_pc) {
     sd(R0, Address(thread, JavaThread::last_Java_pc_offset()));
   }
 }
 // Write serialization page so VM thread can do a pseudo remote membar.
 // We use the current thread pointer to calculate a thread specific
 // offset to write to within the page. This minimizes bus traffic
 // due to cache line collision.
 void MacroAssembler::serialize_memory(Register thread, Register tmp) {
   move(tmp, thread);
   srl(tmp, tmp,os::get_serialize_page_shift_count());
   move(AT, (os::vm_page_size() - sizeof(int)));
   andr(tmp, tmp,AT);
   sw(tmp,Address(tmp, (intptr_t)os::get_memory_serialize_page()));
 }
 // Calls to C land
 //
 // When entering C land, the rbp, & rsp of the last Java frame have to be recorded
 // in the (thread-local) JavaThread object. When leaving C land, the last Java fp
 // has to be reset to 0. This is required to allow proper stack traversal.
 void MacroAssembler::set_last_Java_frame(Register java_thread,
                                          Register last_java_sp,
                                          Register last_java_fp,
                                          address  last_java_pc) {
   // determine java_thread register
   if (!java_thread->is_valid()) {
 #ifndef OPT_THREAD
     java_thread = T2;
     get_thread(java_thread);
 #else
     java_thread = TREG;
 #endif
   }
   // determine last_java_sp register
   if (!last_java_sp->is_valid()) {
     last_java_sp = SP;
   }
   // last_java_fp is optional
   if (last_java_fp->is_valid()) {
     st_ptr(last_java_fp, java_thread, in_bytes(JavaThread::last_Java_fp_offset()));
   }
   // last_java_pc is optional
   if (last_java_pc != NULL) {
     relocate(relocInfo::internal_pc_type);
     li48(AT, (long)last_java_pc);
     st_ptr(AT, java_thread, in_bytes(JavaThread::last_Java_pc_offset()));
   }
   st_ptr(last_java_sp, java_thread, in_bytes(JavaThread::last_Java_sp_offset()));
 }
 void MacroAssembler::set_last_Java_frame(Register last_java_sp,
                                          Register last_java_fp,
                                          address  last_java_pc) {
   // determine last_java_sp register
   if (!last_java_sp->is_valid()) {
     last_java_sp = SP;
   }
   Register thread = TREG;
 #ifndef OPT_THREAD
   get_thread(thread);
 #endif
   // last_java_fp is optional
   if (last_java_fp->is_valid()) {
     sd(last_java_fp, Address(thread, JavaThread::last_Java_fp_offset()));
   }
   // last_java_pc is optional
   if (last_java_pc != NULL) {
     Address java_pc(thread,
                     JavaThread::frame_anchor_offset() + JavaFrameAnchor::last_Java_pc_offset());
     li(AT, (intptr_t)(last_java_pc));
     sd(AT, java_pc);
   }
   sd(last_java_sp, Address(thread, JavaThread::last_Java_sp_offset()));
 }
 //////////////////////////////////////////////////////////////////////////////////
 #ifndef SERIALGC
 void MacroAssembler::g1_write_barrier_pre(Register obj,
 #ifndef _LP64
                                           Register thread,
 #endif
                                           Register tmp,
                                           Register tmp2,
                                           bool tosca_live) {
 /*  LP64_ONLY(Register thread = r15_thread;)
   Address in_progress(thread, in_bytes(JavaThread::satb_mark_queue_offset() +
                                        PtrQueue::byte_offset_of_active()));
   Address index(thread, in_bytes(JavaThread::satb_mark_queue_offset() +
                                        PtrQueue::byte_offset_of_index()));
   Address buffer(thread, in_bytes(JavaThread::satb_mark_queue_offset() +
                                        PtrQueue::byte_offset_of_buf()));
   Label done;
   Label runtime;
   // if (!marking_in_progress) goto done;
   if (in_bytes(PtrQueue::byte_width_of_active()) == 4) {
     cmpl(in_progress, 0);
   } else {
     assert(in_bytes(PtrQueue::byte_width_of_active()) == 1, "Assumption");
     cmpb(in_progress, 0);
   }
   jcc(Assembler::equal, done);
   // if (x.f == NULL) goto done;
   cmpptr(Address(obj, 0), NULL_WORD);
   jcc(Assembler::equal, done);
   // Can we store original value in the thread's buffer?
   LP64_ONLY(movslq(tmp, index);)
   movptr(tmp2, Address(obj, 0));
 #ifdef _LP64
   cmpq(tmp, 0);
 #else
   cmpl(index, 0);
 #endif
   jcc(Assembler::equal, runtime);
 #ifdef _LP64
   subq(tmp, wordSize);
   movl(index, tmp);
   addq(tmp, buffer);
 #else
   subl(index, wordSize);
   movl(tmp, buffer);
   addl(tmp, index);
 #endif
   movptr(Address(tmp, 0), tmp2);
   jmp(done);
   bind(runtime);
   // save the live input values
   if(tosca_live) push(rax);
   push(obj);
 #ifdef _LP64
   movq(c_rarg0, Address(obj, 0));
   call_VM_leaf(CAST_FROM_FN_PTR(address, SharedRuntime::g1_wb_pre), c_rarg0, r15_thread);
 #else
   push(thread);
   call_VM_leaf(CAST_FROM_FN_PTR(address, SharedRuntime::g1_wb_pre), tmp2, thread);
   pop(thread);
 #endif
   pop(obj);
   if(tosca_live) pop(rax);
   bind(done);
 */
 }
 void MacroAssembler::g1_write_barrier_post(Register store_addr,
                                            Register new_val,
 #ifndef _LP64
                                            Register thread,
 #endif
                                            Register tmp,
                                            Register tmp2) {
   /*LP64_ONLY(Register thread = r15_thread;)
   Address queue_index(thread, in_bytes(JavaThread::dirty_card_queue_offset() +
                                        PtrQueue::byte_offset_of_index()));
   Address buffer(thread, in_bytes(JavaThread::dirty_card_queue_offset() +
                                        PtrQueue::byte_offset_of_buf()));
   BarrierSet* bs = Universe::heap()->barrier_set();
   CardTableModRefBS* ct = (CardTableModRefBS*)bs;
   Label done;
   Label runtime;
   // Does store cross heap regions?
   movptr(tmp, store_addr);
   xorptr(tmp, new_val);
   shrptr(tmp, HeapRegion::LogOfHRGrainBytes);
   jcc(Assembler::equal, done);
   // crosses regions, storing NULL?
   cmpptr(new_val, (int32_t) NULL_WORD);
   jcc(Assembler::equal, done);
   // storing region crossing non-NULL, is card already dirty?
   ExternalAddress cardtable((address) ct->byte_map_base);
   assert(sizeof(*ct->byte_map_base) == sizeof(jbyte), "adjust this code");
 #ifdef _LP64
   const Register card_addr = tmp;
   movq(card_addr, store_addr);
   shrq(card_addr, CardTableModRefBS::card_shift);
   lea(tmp2, cardtable);
   // get the address of the card
   addq(card_addr, tmp2);
 #else
   const Register card_index = tmp;
   movl(card_index, store_addr);
   shrl(card_index, CardTableModRefBS::card_shift);
   Address index(noreg, card_index, Address::times_1);
   const Register card_addr = tmp;
   lea(card_addr, as_Address(ArrayAddress(cardtable, index)));
 #endif
   cmpb(Address(card_addr, 0), 0);
   jcc(Assembler::equal, done);
   // storing a region crossing, non-NULL oop, card is clean.
   // dirty card and log.
   movb(Address(card_addr, 0), 0);
   cmpl(queue_index, 0);
   jcc(Assembler::equal, runtime);
   subl(queue_index, wordSize);
   movptr(tmp2, buffer);
 #ifdef _LP64
   movslq(rscratch1, queue_index);
   addq(tmp2, rscratch1);
   movq(Address(tmp2, 0), card_addr);
 #else
   addl(tmp2, queue_index);
   movl(Address(tmp2, 0), card_index);
 #endif
   jmp(done);
   bind(runtime);
   // save the live input values
   push(store_addr);
   push(new_val);
 #ifdef _LP64
   call_VM_leaf(CAST_FROM_FN_PTR(address, SharedRuntime::g1_wb_post), card_addr, r15_thread);
 #else
   push(thread);
   call_VM_leaf(CAST_FROM_FN_PTR(address, SharedRuntime::g1_wb_post), card_addr, thread);
   pop(thread);
 #endif
   pop(new_val);
   pop(store_addr);
   bind(done);
 */
 }
 #endif // SERIALGC
 //////////////////////////////////////////////////////////////////////////////////
 void MacroAssembler::store_check(Register obj) {
   // Does a store check for the oop in register obj. The content of
   // register obj is destroyed afterwards.
   store_check_part_1(obj);
   store_check_part_2(obj);
 }
 void MacroAssembler::store_check(Register obj, Address dst) {
   store_check(obj);
 }
 // split the store check operation so that other instructions can be scheduled inbetween
 void MacroAssembler::store_check_part_1(Register obj) {
   BarrierSet* bs = Universe::heap()->barrier_set();
   assert(bs->kind() == BarrierSet::CardTableModRef, "Wrong barrier set kind");
 #ifdef _LP64
   dsrl(obj, obj, CardTableModRefBS::card_shift);
 #else
   shr(obj, CardTableModRefBS::card_shift);
 #endif
 }
 void MacroAssembler::store_check_part_2(Register obj) {
   BarrierSet* bs = Universe::heap()->barrier_set();
   assert(bs->kind() == BarrierSet::CardTableModRef, "Wrong barrier set kind");
   CardTableModRefBS* ct = (CardTableModRefBS*)bs;
   assert(sizeof(*ct->byte_map_base) == sizeof(jbyte), "adjust this code");
   li(AT, (long)ct->byte_map_base);
 #ifdef _LP64
   dadd(AT, AT, obj);
 #else
   add(AT, AT, obj);
 #endif
   sb(R0, AT, 0);
 }
 /*
 void MacroAssembler::subptr(Register dst, int32_t imm32) {
   LP64_ONLY(subq(dst, imm32)) NOT_LP64(subl(dst, imm32));
 }
 void MacroAssembler::subptr(Register dst, Register src) {
   LP64_ONLY(subq(dst, src)) NOT_LP64(subl(dst, src));
 }
 void MacroAssembler::test32(Register src1, AddressLiteral src2) {
   // src2 must be rval
   if (reachable(src2)) {
     testl(src1, as_Address(src2));
   } else {
     lea(rscratch1, src2);
     testl(src1, Address(rscratch1, 0));
   }
 }
 // C++ bool manipulation
 void MacroAssembler::testbool(Register dst) {
   if(sizeof(bool) == 1)
     testb(dst, 0xff);
   else if(sizeof(bool) == 2) {
     // testw implementation needed for two byte bools
     ShouldNotReachHere();
   } else if(sizeof(bool) == 4)
     testl(dst, dst);
   else
     // unsupported
     ShouldNotReachHere();
 }
 void MacroAssembler::testptr(Register dst, Register src) {
   LP64_ONLY(testq(dst, src)) NOT_LP64(testl(dst, src));
 }
 */
 // Defines obj, preserves var_size_in_bytes, okay for t2 == var_size_in_bytes.
 void MacroAssembler::tlab_allocate(Register obj, Register var_size_in_bytes, int con_size_in_bytes,
                                    Register t1, Register t2, Label& slow_case) {
   assert_different_registers(obj, var_size_in_bytes, t1, t2, AT);
   Register end = t2;
 #ifndef OPT_THREAD
   Register thread = t1;
   get_thread(thread);
 #else
   Register thread = TREG;
 #endif
   verify_tlab(t1, t2);//blows t1&t2
   ld_ptr(obj, thread, in_bytes(JavaThread::tlab_top_offset()));
   if (var_size_in_bytes == NOREG) {
     // i dont think we need move con_size_in_bytes to a register first.
     // by yjl 8/17/2005
     assert(is_simm16(con_size_in_bytes), "fixme by moving imm to a register first");
     addi(end, obj, con_size_in_bytes);
   } else {
     add(end, obj, var_size_in_bytes);
   }
   ld_ptr(AT, thread, in_bytes(JavaThread::tlab_end_offset()));
   sltu(AT, AT, end);
   bne_far(AT, R0, slow_case);
   delayed()->nop();
   // update the tlab top pointer
   st_ptr(end, thread, in_bytes(JavaThread::tlab_top_offset()));
   // recover var_size_in_bytes if necessary
   /*if (var_size_in_bytes == end) {
     sub(var_size_in_bytes, end, obj);
     }*/
   verify_tlab(t1, t2);
 }
 // Defines obj, preserves var_size_in_bytes
 void MacroAssembler::eden_allocate(Register obj, Register var_size_in_bytes, int con_size_in_bytes,
 		Register t1, Register t2, Label& slow_case) {
   assert_different_registers(obj, var_size_in_bytes, t1, AT);
   if (CMSIncrementalMode || !Universe::heap()->supports_inline_contig_alloc()) { //by yyq
     // No allocation in the shared eden.
     b_far(slow_case);
     delayed()->nop();
   } else {
 #ifndef _LP64
     Address heap_top(t1, Assembler::split_low((intptr_t)Universe::heap()->top_addr()));
     lui(t1, split_high((intptr_t)Universe::heap()->top_addr()));
 #else
     Address heap_top(t1);
     li(t1, (long)Universe::heap()->top_addr());
 #endif
     ld_ptr(obj, heap_top);
     Register end = t2;
     Label retry;
     bind(retry);
     if (var_size_in_bytes == NOREG) {
     // i dont think we need move con_size_in_bytes to a register first.
     // by yjl 8/17/2005
       assert(is_simm16(con_size_in_bytes), "fixme by moving imm to a register first");
       addi(end, obj, con_size_in_bytes);
     } else {
       add(end, obj, var_size_in_bytes);
     }
     // if end < obj then we wrapped around => object too long => slow case
     sltu(AT, end, obj);
     bne_far(AT, R0, slow_case);
     delayed()->nop();
     //lui(AT, split_high((int)Universe::heap()->end_addr()));
     //lw(AT, AT, split_low((int)Universe::heap()->end_addr()));
     li(AT, (long)Universe::heap()->end_addr());
     sltu(AT, AT, end);
     bne_far(AT, R0, slow_case);
     delayed()->nop();
     // Compare obj with the top addr, and if still equal, store the new top addr in
     // end at the address of the top addr pointer. Sets ZF if was equal, and clears
     // it otherwise. Use lock prefix for atomicity on MPs.
     if (os::is_MP()) {
     	///lock();
     }
     // if someone beat us on the allocation, try again, otherwise continue
     cmpxchg(end, heap_top, obj);
     beq_far(AT, R0, retry);    //by yyq
     delayed()->nop();
   }
 }
 void MacroAssembler::tlab_refill(Label& retry, Label& try_eden, Label& slow_case) {
 	Register top = T0;
 	Register t1  = T1;
 /* Jin: tlab_refill() is called in
      [c1_Runtime1_mips.cpp] Runtime1::generate_code_for(new_type_array_id);
   In generate_code_for(), T2 has been assigned as a register(length), which is used
  after calling tlab_refill();
   Therefore, tlab_refill() should not use T2.
  Source:
 Exception in thread "main" java.lang.ArrayIndexOutOfBoundsException
         at java.lang.System.arraycopy(Native Method)
         at java.util.Arrays.copyOf(Arrays.java:2799)	<-- alloc_array
         at sun.misc.Resource.getBytes(Resource.java:117)
         at java.net.URLClassLoader.defineClass(URLClassLoader.java:273)
         at java.net.URLClassLoader.findClass(URLClassLoader.java:205)
         at java.lang.ClassLoader.loadClass(ClassLoader.java:321)
  */
 	Register t2  = T9;
 	Register t3  = T3;
 	Register thread_reg = T8;
 	Label do_refill, discard_tlab;
 	if (CMSIncrementalMode || !Universe::heap()->supports_inline_contig_alloc()) { //by yyq
 		// No allocation in the shared eden.
 		b(slow_case);
 		delayed()->nop();
 	}
 	get_thread(thread_reg);
 	ld_ptr(top, thread_reg, in_bytes(JavaThread::tlab_top_offset()));
 	ld_ptr(t1, thread_reg, in_bytes(JavaThread::tlab_end_offset()));
 	// calculate amount of free space
 	sub(t1, t1, top);
 	shr(t1, LogHeapWordSize);
 	// Retain tlab and allocate object in shared space if
 	// the amount free in the tlab is too large to discard.
 	ld_ptr(t2, thread_reg, in_bytes(JavaThread::tlab_refill_waste_limit_offset()));
 	slt(AT, t2, t1);
 	beq(AT, R0, discard_tlab);
 	delayed()->nop();
 	// Retain
 #ifndef _LP64
 	move(AT, ThreadLocalAllocBuffer::refill_waste_limit_increment());
 #else
 	li(AT, ThreadLocalAllocBuffer::refill_waste_limit_increment());
 #endif
 	add(t2, t2, AT);
 	st_ptr(t2, thread_reg, in_bytes(JavaThread::tlab_refill_waste_limit_offset()));
 	if (TLABStats) {
 		// increment number of slow_allocations
 		lw(AT, thread_reg, in_bytes(JavaThread::tlab_slow_allocations_offset()));
 		addiu(AT, AT, 1);
 		sw(AT, thread_reg, in_bytes(JavaThread::tlab_slow_allocations_offset()));
 	}
 	b(try_eden);
 	delayed()->nop();
   bind(discard_tlab);
 	if (TLABStats) {
 		// increment number of refills
 		lw(AT, thread_reg, in_bytes(JavaThread::tlab_number_of_refills_offset()));
 		addi(AT, AT, 1);
 		sw(AT, thread_reg, in_bytes(JavaThread::tlab_number_of_refills_offset()));
 		// accumulate wastage -- t1 is amount free in tlab
 		lw(AT, thread_reg, in_bytes(JavaThread::tlab_fast_refill_waste_offset()));
 		add(AT, AT, t1);
 		sw(AT, thread_reg, in_bytes(JavaThread::tlab_fast_refill_waste_offset()));
 	}
 	// if tlab is currently allocated (top or end != null) then
 	// fill [top, end + alignment_reserve) with array object
 	beq(top, R0, do_refill);
 	delayed()->nop();
 	// set up the mark word
 	li(AT, (long)markOopDesc::prototype()->copy_set_hash(0x2));
 	st_ptr(AT, top, oopDesc::mark_offset_in_bytes());
 	// set the length to the remaining space
 	addi(t1, t1, - typeArrayOopDesc::header_size(T_INT));
 	addi(t1, t1, ThreadLocalAllocBuffer::alignment_reserve());
 	shl(t1, log2_intptr(HeapWordSize/sizeof(jint)));
 	sw(t1, top, arrayOopDesc::length_offset_in_bytes());
 	// set klass to intArrayKlass
 #ifndef _LP64
 	lui(AT, split_high((intptr_t)Universe::intArrayKlassObj_addr()));
 	lw(t1, AT, split_low((intptr_t)Universe::intArrayKlassObj_addr()));
 #else
 	li(AT, (intptr_t)Universe::intArrayKlassObj_addr());
 	ld_ptr(t1, AT, 0);
 #endif
 	//st_ptr(t1, top, oopDesc::klass_offset_in_bytes());
 	store_klass(top, t1);
 	// refill the tlab with an eden allocation
 	bind(do_refill);
 	ld_ptr(t1, thread_reg, in_bytes(JavaThread::tlab_size_offset()));
 	shl(t1, LogHeapWordSize);
 	// add object_size ??
 	eden_allocate(top, t1, 0, t2, t3, slow_case);
 	// Check that t1 was preserved in eden_allocate.
 #ifdef ASSERT
 	if (UseTLAB) {
 		Label ok;
 		assert_different_registers(thread_reg, t1);
 		ld_ptr(AT, thread_reg, in_bytes(JavaThread::tlab_size_offset()));
 		shl(AT, LogHeapWordSize);
 		beq(AT, t1, ok);
 		delayed()->nop();
 		stop("assert(t1 != tlab size)");
 		should_not_reach_here();
 		bind(ok);
 	}
 #endif
 	st_ptr(top, thread_reg, in_bytes(JavaThread::tlab_start_offset()));
 	st_ptr(top, thread_reg, in_bytes(JavaThread::tlab_top_offset()));
 	add(top, top, t1);
 	addi(top, top, - ThreadLocalAllocBuffer::alignment_reserve_in_bytes());
 	st_ptr(top, thread_reg, in_bytes(JavaThread::tlab_end_offset()));
 	verify_tlab(t1, t2);
 	b(retry);
 	delayed()->nop();
 }
 static const double     pi_4 =  0.7853981633974483;
 // the x86 version is to clumsy, i dont think we need that fuss. maybe i'm wrong, FIXME
 // must get argument(a double) in F12/F13
 //void MacroAssembler::trigfunc(char trig, bool preserve_cpu_regs, int num_fpu_regs_in_use) {
 //We need to preseve the register which maybe modified during the Call @Jerome
 void MacroAssembler::trigfunc(char trig, int num_fpu_regs_in_use) {
 //save all modified register here
 //	if (preserve_cpu_regs) {
 //	}
 //FIXME, in the disassembly of tirgfunc, only used V0,V1,T9, SP,RA,so we ony save V0,V1,T9
 	pushad();
 //we should preserve the stack space before we call
 	addi(SP, SP, -wordSize * 2);
         switch (trig){
 		case 's' :
               		call( CAST_FROM_FN_PTR(address, SharedRuntime::dsin), relocInfo::runtime_call_type );
 			delayed()->nop();
 			break;
 		case 'c':
 			call( CAST_FROM_FN_PTR(address, SharedRuntime::dcos), relocInfo::runtime_call_type );
 			delayed()->nop();
 			break;
 		case 't':
 			call( CAST_FROM_FN_PTR(address, SharedRuntime::dtan), relocInfo::runtime_call_type );
 			delayed()->nop();
 			break;
 		default:assert (false, "bad intrinsic");
 		break;
 	}
 	addi(SP, SP, wordSize * 2);
 	popad();
 //	if (preserve_cpu_regs) {
 //	}
 }
 /*
 void MacroAssembler::ucomisd(XMMRegister dst, AddressLiteral src) {
   ucomisd(dst, as_Address(src));
 }
 void MacroAssembler::ucomiss(XMMRegister dst, AddressLiteral src) {
   ucomiss(dst, as_Address(src));
 }
 void MacroAssembler::xorpd(XMMRegister dst, AddressLiteral src) {
   if (reachable(src)) {
     xorpd(dst, as_Address(src));
   } else {
     lea(rscratch1, src);
     xorpd(dst, Address(rscratch1, 0));
   }
 }
 void MacroAssembler::xorps(XMMRegister dst, AddressLiteral src) {
   if (reachable(src)) {
     xorps(dst, as_Address(src));
   } else {
     lea(rscratch1, src);
     xorps(dst, Address(rscratch1, 0));
   }
 }
  */
 #ifdef _LP64
 void MacroAssembler::li(Register rd, long imm) {
   if (imm <= max_jint && imm >= min_jint) {
     li32(rd, (int)imm);
   } else if (julong(imm) <= 0xFFFFFFFF) {
     assert_not_delayed();
     // lui sign-extends, so we can't use that.
     ori(rd, R0, julong(imm) >> 16);
     dsll(rd, rd, 16);
     ori(rd, rd, split_low(imm));
   //aoqi_test
   //} else if ((imm > 0) && ((imm >> 48) == 0)) {
   } else if ((imm > 0) && is_simm16(imm >> 32)) {
     /* A 48-bit address */
     li48(rd, imm);
   } else {
     li64(rd, imm);
   }
 }
 #else
 void MacroAssembler::li(Register rd, long imm) {
   li32(rd, (int)imm);
 }
 #endif
 void MacroAssembler::li32(Register reg, int imm) {
   if (is_simm16(imm)) {
     /* Jin: for imm < 0, we should use addi instead of addiu.
      *
      *  java.lang.StringCoding$StringDecoder.decode(jobject, jint, jint)
      *
      *  78 move [int:-1|I] [a0|I]
      *    : daddi a0, zero, 0xffffffff  (correct)
      *    : daddiu a0, zero, 0xffffffff (incorrect)
      */
     if (imm >= 0)
       addiu(reg, R0, imm);
     else
       addi(reg, R0, imm);
   } else {
     lui(reg, split_low(imm >> 16));
     if (split_low(imm))
       ori(reg, reg, split_low(imm));
   }
 }
 #ifdef _LP64
 void MacroAssembler::li64(Register rd, long imm) {
   assert_not_delayed();
   lui(rd, imm >> 48);
   ori(rd, rd, split_low(imm >> 32));
   dsll(rd, rd, 16);
   ori(rd, rd, split_low(imm >> 16));
   dsll(rd, rd, 16);
   ori(rd, rd, split_low(imm));
 }
 void MacroAssembler::li48(Register rd, long imm) {
   assert(is_simm16(imm >> 32), "Not a 48-bit address");
   lui(rd, imm >> 32);
   ori(rd, rd, split_low(imm >> 16));
   dsll(rd, rd, 16);
   ori(rd, rd, split_low(imm));
 }
 #endif
 // NOTE: i dont push eax as i486.
 // the x86 save eax for it use eax as the jump register
 void MacroAssembler::verify_oop(Register reg, const char* s) {
   /*
      if (!VerifyOops) return;
   // Pass register number to verify_oop_subroutine
   char* b = new char[strlen(s) + 50];
   sprintf(b, "verify_oop: %s: %s", reg->name(), s);
   push(rax);                          // save rax,
   push(reg);                          // pass register argument
   ExternalAddress buffer((address) b);
   // avoid using pushptr, as it modifies scratch registers
   // and our contract is not to modify anything
   movptr(rax, buffer.addr());
   push(rax);
   // call indirectly to solve generation ordering problem
   movptr(rax, ExternalAddress(StubRoutines::verify_oop_subroutine_entry_address()));
   call(rax);
    */
   if (!VerifyOops) return;
   const char * b = NULL;
   stringStream ss;
   ss.print("verify_oop: %s: %s", reg->name(), s);
   b = code_string(ss.as_string());
 #ifdef _LP64
   pushad();
   move(A1, reg);
   li(A0, (long)b);
   li(AT, (long)StubRoutines::verify_oop_subroutine_entry_address());
   ld(T9, AT, 0);
   jalr(T9);
   delayed()->nop();
   popad();
 #else
   // Pass register number to verify_oop_subroutine
   sw(T0, SP, - wordSize);
   sw(T1, SP, - 2*wordSize);
   sw(RA, SP, - 3*wordSize);
   sw(A0, SP ,- 4*wordSize);
   sw(A1, SP ,- 5*wordSize);
   sw(AT, SP ,- 6*wordSize);
   sw(T9, SP ,- 7*wordSize);
   addiu(SP, SP, - 7 * wordSize);
   move(A1, reg);
   li(A0, (long)b);
   // call indirectly to solve generation ordering problem
   li(AT, (long)StubRoutines::verify_oop_subroutine_entry_address());
   lw(T9, AT, 0);
   jalr(T9);
   delayed()->nop();
   lw(T0, SP, 6* wordSize);
   lw(T1, SP, 5* wordSize);
   lw(RA, SP, 4* wordSize);
   lw(A0, SP, 3* wordSize);
   lw(A1, SP, 2* wordSize);
   lw(AT, SP, 1* wordSize);
   lw(T9, SP, 0* wordSize);
   addiu(SP, SP, 7 * wordSize);
 #endif
 }
 void MacroAssembler::verify_oop_addr(Address addr, const char* s) {
 	if (!VerifyOops) {
 		nop();
 		return;
 	}
 	// Pass register number to verify_oop_subroutine
 	const char * b = NULL;
 	stringStream ss;
 	ss.print("verify_oop_addr: %s",  s);
 	b = code_string(ss.as_string());
 	st_ptr(T0, SP, - wordSize);
 	st_ptr(T1, SP, - 2*wordSize);
 	st_ptr(RA, SP, - 3*wordSize);
 	st_ptr(A0, SP, - 4*wordSize);
 	st_ptr(A1, SP, - 5*wordSize);
 	st_ptr(AT, SP, - 6*wordSize);
 	st_ptr(T9, SP, - 7*wordSize);
 	ld_ptr(A1, addr);   // addr may use SP, so load from it before change SP
 	addiu(SP, SP, - 7 * wordSize);
 	li(A0, (long)b);
 	// call indirectly to solve generation ordering problem
 	li(AT, (long)StubRoutines::verify_oop_subroutine_entry_address());
 	ld_ptr(T9, AT, 0);
 	jalr(T9);
 	delayed()->nop();
 	ld_ptr(T0, SP, 6* wordSize);
 	ld_ptr(T1, SP, 5* wordSize);
 	ld_ptr(RA, SP, 4* wordSize);
 	ld_ptr(A0, SP, 3* wordSize);
 	ld_ptr(A1, SP, 2* wordSize);
 	ld_ptr(AT, SP, 1* wordSize);
 	ld_ptr(T9, SP, 0* wordSize);
 	addiu(SP, SP, 7 * wordSize);
 }
 // used registers :  T0, T1
 void MacroAssembler::verify_oop_subroutine() {
   // RA: ra
   // A0: char* error message
   // A1: oop   object to verify
   Label exit, error;
   // increment counter
   li(T0, (long)StubRoutines::verify_oop_count_addr());
   lw(AT, T0, 0);
 #ifdef _LP64
 //FIXME, aoqi: rewrite addi, addu, etc in 64bits mode.
   daddi(AT, AT, 1);
 #else
   addi(AT, AT, 1);
 #endif
   sw(AT, T0, 0);
   // make sure object is 'reasonable'
   beq(A1, R0, exit);         // if obj is NULL it is ok
   delayed()->nop();
   // Check if the oop is in the right area of memory
   //const int oop_mask = Universe::verify_oop_mask();
   //const int oop_bits = Universe::verify_oop_bits();
   const uintptr_t oop_mask = Universe::verify_oop_mask();
   const uintptr_t oop_bits = Universe::verify_oop_bits();
   li(AT, oop_mask);
   andr(T0, A1, AT);
   li(AT, oop_bits);
   bne(T0, AT, error);
   delayed()->nop();
   // make sure klass is 'reasonable'
   //add for compressedoops
   reinit_heapbase();
   //add for compressedoops
   load_klass(T0, A1);
   beq(T0, R0, error);                        // if klass is NULL it is broken
   delayed()->nop();
   #if 0
   //FIXME:wuhui.
   // Check if the klass is in the right area of memory
   //const int klass_mask = Universe::verify_klass_mask();
   //const int klass_bits = Universe::verify_klass_bits();
   const uintptr_t klass_mask = Universe::verify_klass_mask();
   const uintptr_t klass_bits = Universe::verify_klass_bits();
   li(AT, klass_mask);
   andr(T1, T0, AT);
   li(AT, klass_bits);
   bne(T1, AT, error);
   delayed()->nop();
   // make sure klass' klass is 'reasonable'
   //add for compressedoops
   load_klass(T0, T0);
   beq(T0, R0, error);  // if klass' klass is NULL it is broken
   delayed()->nop();
   li(AT, klass_mask);
   andr(T1, T0, AT);
   li(AT, klass_bits);
   bne(T1, AT, error);
   delayed()->nop();     // if klass not in right area of memory it is broken too.
 #endif
   // return if everything seems ok
   bind(exit);
   jr(RA);
   delayed()->nop();
   // handle errors
   bind(error);
   pushad();
 #ifndef _LP64
   addi(SP, SP, (-1) * wordSize);
 #endif
   call(CAST_FROM_FN_PTR(address, MacroAssembler::debug), relocInfo::runtime_call_type);
   delayed()->nop();
 #ifndef _LP64
   addiu(SP, SP, 1 * wordSize);
 #endif
   popad();
   jr(RA);
   delayed()->nop();
 }
 void MacroAssembler::verify_tlab(Register t1, Register t2) {
 #ifdef ASSERT
   assert_different_registers(t1, t2, AT);
   if (UseTLAB && VerifyOops) {
     Label next, ok;
     get_thread(t1);
     ld_ptr(t2, t1, in_bytes(JavaThread::tlab_top_offset()));
     ld_ptr(AT, t1, in_bytes(JavaThread::tlab_start_offset()));
     sltu(AT, t2, AT);
     beq(AT, R0, next);
     delayed()->nop();
     stop("assert(top >= start)");
     bind(next);
     ld_ptr(AT, t1, in_bytes(JavaThread::tlab_end_offset()));
     sltu(AT, AT, t2);
     beq(AT, R0, ok);
     delayed()->nop();
     stop("assert(top <= end)");
     bind(ok);
     /*
        Label next, ok;
        Register t1 = rsi;
        Register thread_reg = NOT_LP64(rbx) LP64_ONLY(r15_thread);
        push(t1);
        NOT_LP64(push(thread_reg));
        NOT_LP64(get_thread(thread_reg));
        movptr(t1, Address(thread_reg, in_bytes(JavaThread::tlab_top_offset())));
        cmpptr(t1, Address(thread_reg, in_bytes(JavaThread::tlab_start_offset())));
        jcc(Assembler::aboveEqual, next);
        stop("assert(top >= start)");
        should_not_reach_here();
        bind(next);
        movptr(t1, Address(thread_reg, in_bytes(JavaThread::tlab_end_offset())));
        cmpptr(t1, Address(thread_reg, in_bytes(JavaThread::tlab_top_offset())));
        jcc(Assembler::aboveEqual, ok);
        stop("assert(top <= end)");
        should_not_reach_here();
        bind(ok);
        NOT_LP64(pop(thread_reg));
        pop(t1);
      */
   }
 #endif
 }
  RegisterOrConstant MacroAssembler::delayed_value_impl(intptr_t* delayed_value_addr,
                                                        Register tmp,
                                                        int offset) {
    intptr_t value = *delayed_value_addr;
    if (value != 0)
    return RegisterOrConstant(value + offset);
    AddressLiteral a(delayed_value_addr);
    // load indirectly to solve generation ordering problem
    //movptr(tmp, ExternalAddress((address) delayed_value_addr));
    //ld(tmp, a);
   /* #ifdef ASSERT
    { Label L;
      testptr(tmp, tmp);
      if (WizardMode) {
             jcc(Assembler::notZero, L);
             char* buf = new char[40];
             sprintf(buf, "DelayedValue="INTPTR_FORMAT, delayed_value_addr[1]);
             STOP(buf);
                       } else {
             jccb(Assembler::notZero, L);
             hlt();
                              }
      bind(L);
    }
    #endif*/
    if (offset != 0)
      daddi(tmp,tmp, offset);
    return RegisterOrConstant(tmp);
  }
 void MacroAssembler::hswap(Register reg) {
   //andi(reg, reg, 0xffff);
   srl(AT, reg, 8);
   sll(reg, reg, 24);
   sra(reg, reg, 16);
   orr(reg, reg, AT);
 }
 void MacroAssembler::huswap(Register reg) {
 #ifdef _LP64
   dsrl(AT, reg, 8);
   dsll(reg, reg, 24);
   dsrl(reg, reg, 16);
   orr(reg, reg, AT);
   andi(reg, reg, 0xffff);
 #else
   //andi(reg, reg, 0xffff);
   srl(AT, reg, 8);
   sll(reg, reg, 24);
   srl(reg, reg, 16);
   orr(reg, reg, AT);
 #endif
 }
 // something funny to do this will only one more register AT
 // by yjl 6/29/2005
 void MacroAssembler::swap(Register reg) {
 	srl(AT, reg, 8);
 	sll(reg, reg, 24);
 	orr(reg, reg, AT);
 	//reg : 4 1 2 3
 	srl(AT, AT, 16);
 	xorr(AT, AT, reg);
 	andi(AT, AT, 0xff);
 	//AT : 0 0 0 1^3);
 	xorr(reg, reg, AT);
 	//reg : 4 1 2 1
 	sll(AT, AT, 16);
 	xorr(reg, reg, AT);
 	//reg : 4 3 2 1
 }
 #ifdef _LP64
 /* do 32-bit CAS using MIPS64 lld/scd
   Jin: cas_int should only compare 32-bits of the memory value.
        However, lld/scd will do 64-bit operation, which violates the intention of cas_int.
        To simulate a 32-bit atomic operation, the value loaded with LLD should be split into
        tow halves, and only the low-32 bits is compared. If equals, the low-32 bits of newval,
        plus the high-32 bits or memory value, are stored togethor with SCD.
 Example:
       double d = 3.1415926;
       System.err.println("hello" + d);
   sun.misc.FloatingDecimal$1.<init>()
    |
    `- java.util.concurrent.atomic.AtomicInteger::compareAndSet()
 cas_int [a7a7|J] [a0|I] [a6|I]
 // a0: 0xffffffffe8ea9f63 pc: 0x55647f3354
 // a6: 0x4ab325aa
 again:
 x00000055647f3c5c: lld at, 0x0(a7)                          ; 64-bit load, "0xe8ea9f63"
 x00000055647f3c60: sll t9, at, 0                            ; t9: low-32 bits (sign extended)
 x00000055647f3c64: dsrl32 t8, at, 0                         ; t8: high-32 bits
 x00000055647f3c68: dsll32 t8, t8, 0
 x00000055647f3c6c: bne t9, a0, 0x00000055647f3c9c           ; goto nequal
 x00000055647f3c70: sll zero, zero, 0
 x00000055647f3c74: ori v1, zero, 0xffffffff                 ; v1: low-32 bits of newval (sign unextended)
 x00000055647f3c78: dsll v1, v1, 16                          ; v1 = a6 & 0xFFFFFFFF;
 x00000055647f3c7c: ori v1, v1, 0xffffffff
 x00000055647f3c80: and v1, a6, v1
 x00000055647f3c84: or at, t8, v1
 x00000055647f3c88: scd at, 0x0(a7)
 x00000055647f3c8c: beq at, zero, 0x00000055647f3c5c         ; goto again
 x00000055647f3c90: sll zero, zero, 0
 x00000055647f3c94: beq zero, zero, 0x00000055647f45ac       ; goto done
 x00000055647f3c98: sll zero, zero, 0
 nequal:
 x00000055647f45a4: dadd a0, t9, zero
 x00000055647f45a8: dadd at, zero, zero
 done:
 */
 void MacroAssembler::cmpxchg32(Register x_reg, Address dest, Register c_reg) {
 #if 0
   Label done, again, nequal;
   bind(again);
   sync();
   lld(AT, dest);
   /* T9:  32 bits, sign extended
    * V1: low 32 bits, sign unextended
    * T8: high 32 bits (may be another variables's space)
    */
   sll(T9, AT, 0);	// Use 32-bit sll to extend bit 31
   dsrl32(T8, AT, 0);
   dsll32(T8, T8, 0);
   bne(T9, c_reg, nequal);
   delayed()->nop();
   ori(V1, R0, 0xFFFF);
   dsll(V1, V1, 16);
   ori(V1, V1, 0xFFFF);
   andr(V1, x_reg, V1);
   orr(AT, T8, V1);
   scd(AT, dest);
   beq(AT, R0, again);
   delayed()->nop();
   b(done);
   delayed()->nop();
   // not xchged
   bind(nequal);
   move(c_reg, T9);
   move(AT, R0);
   bind(done);
 #else
   /* 2012/11/11 Jin: MIPS64 can use ll/sc for 32-bit atomic memory access */
   Label done, again, nequal;
   bind(again);
   sync();
   ll(AT, dest);
   bne(AT, c_reg, nequal);
   delayed()->nop();
   move(AT, x_reg);
   sc(AT, dest);
   beq(AT, R0, again);
   delayed()->nop();
   b(done);
   delayed()->nop();
   // not xchged
   bind(nequal);
   sync();
   move(c_reg, AT);
   move(AT, R0);
   bind(done);
 #endif
 }
 #endif	// cmpxchg32
 void MacroAssembler::cmpxchg(Register x_reg, Address dest, Register c_reg) {
   Label done, again, nequal;
   bind(again);
 #ifdef _LP64
   sync();
   lld(AT, dest);
 #else
   sync();
   ll(AT, dest);
 #endif
   bne(AT, c_reg, nequal);
   delayed()->nop();
   move(AT, x_reg);
 #ifdef _LP64
   scd(AT, dest);
 #else
   sc(AT, dest);
 #endif
   beq(AT, R0, again);
   delayed()->nop();
   b(done);
   delayed()->nop();
   // not xchged
   bind(nequal);
   sync();
   move(c_reg, AT);
   move(AT, R0);
   bind(done);
 }
 void MacroAssembler::cmpxchg8(Register x_regLo, Register x_regHi, Address dest, Register c_regLo, Register c_regHi) {
 	Label done, again, nequal;
 	Register x_reg = x_regLo;
 	dsll32(x_regHi, x_regHi, 0);
 	dsll32(x_regLo, x_regLo, 0);
 	dsrl32(x_regLo, x_regLo, 0);
 	orr(x_reg, x_regLo, x_regHi);
 	Register c_reg = c_regLo;
 	dsll32(c_regHi, c_regHi, 0);
 	dsll32(c_regLo, c_regLo, 0);
 	dsrl32(c_regLo, c_regLo, 0);
 	orr(c_reg, c_regLo, c_regHi);
 	bind(again);
 	sync();
 	lld(AT, dest);
 	bne(AT, c_reg, nequal);
 	delayed()->nop();
 	//move(AT, x_reg);
 	dadd(AT, x_reg, R0);
 	scd(AT, dest);
 	beq(AT, R0, again);
 	delayed()->nop();
 	b(done);
 	delayed()->nop();
 	// not xchged
 	bind(nequal);
 	sync();
 	//move(c_reg, AT);
 	//move(AT, R0);
 	dadd(c_reg, AT, R0);
 	dadd(AT, R0, R0);
 	bind(done);
 }
 // be sure the three register is different
 void MacroAssembler::rem_s(FloatRegister fd, FloatRegister fs, FloatRegister ft, FloatRegister tmp) {
   assert_different_registers(tmp, fs, ft);
 	div_s(tmp, fs, ft);
 	trunc_l_s(tmp, tmp);
 	cvt_s_l(tmp, tmp);
 	mul_s(tmp, tmp, ft);
 	sub_s(fd, fs, tmp);
 }
 // be sure the three register is different
 void MacroAssembler::rem_d(FloatRegister fd, FloatRegister fs, FloatRegister ft, FloatRegister tmp) {
 	assert_different_registers(tmp, fs, ft);
 	div_d(tmp, fs, ft);
 	trunc_l_d(tmp, tmp);
 	cvt_d_l(tmp, tmp);
 	mul_d(tmp, tmp, ft);
 	sub_d(fd, fs, tmp);
 }
 // Fast_Lock and Fast_Unlock used by C2
 // Because the transitions from emitted code to the runtime
 // monitorenter/exit helper stubs are so slow it's critical that
 // we inline both the stack-locking fast-path and the inflated fast path.
 //
 // See also: cmpFastLock and cmpFastUnlock.
 //
 // What follows is a specialized inline transliteration of the code
 // in slow_enter() and slow_exit().  If we're concerned about I$ bloat
 // another option would be to emit TrySlowEnter and TrySlowExit methods
 // at startup-time.  These methods would accept arguments as
 // (rax,=Obj, rbx=Self, rcx=box, rdx=Scratch) and return success-failure
 // indications in the icc.ZFlag.  Fast_Lock and Fast_Unlock would simply
 // marshal the arguments and emit calls to TrySlowEnter and TrySlowExit.
 // In practice, however, the # of lock sites is bounded and is usually small.
 // Besides the call overhead, TrySlowEnter and TrySlowExit might suffer
 // if the processor uses simple bimodal branch predictors keyed by EIP
 // Since the helper routines would be called from multiple synchronization
 // sites.
 //
 // An even better approach would be write "MonitorEnter()" and "MonitorExit()"
 // in java - using j.u.c and unsafe - and just bind the lock and unlock sites
 // to those specialized methods.  That'd give us a mostly platform-independent
 // implementation that the JITs could optimize and inline at their pleasure.
 // Done correctly, the only time we'd need to cross to native could would be
 // to park() or unpark() threads.  We'd also need a few more unsafe operators
 // to (a) prevent compiler-JIT reordering of non-volatile accesses, and
 // (b) explicit barriers or fence operations.
 //
 // TODO:
 //
 // *  Arrange for C2 to pass "Self" into Fast_Lock and Fast_Unlock in one of the registers (scr).
 //    This avoids manifesting the Self pointer in the Fast_Lock and Fast_Unlock terminals.
 //    Given TLAB allocation, Self is usually manifested in a register, so passing it into
 //    the lock operators would typically be faster than reifying Self.
 //
 // *  Ideally I'd define the primitives as:
 //       fast_lock   (nax Obj, nax box, EAX tmp, nax scr) where box, tmp and scr are KILLED.
 //       fast_unlock (nax Obj, EAX box, nax tmp) where box and tmp are KILLED
 //    Unfortunately ADLC bugs prevent us from expressing the ideal form.
 //    Instead, we're stuck with a rather awkward and brittle register assignments below.
 //    Furthermore the register assignments are overconstrained, possibly resulting in
 //    sub-optimal code near the synchronization site.
 //
 // *  Eliminate the sp-proximity tests and just use "== Self" tests instead.
 //    Alternately, use a better sp-proximity test.
 //
 // *  Currently ObjectMonitor._Owner can hold either an sp value or a (THREAD *) value.
 //    Either one is sufficient to uniquely identify a thread.
 //    TODO: eliminate use of sp in _owner and use get_thread(tr) instead.
 //
 // *  Intrinsify notify() and notifyAll() for the common cases where the
 //    object is locked by the calling thread but the waitlist is empty.
 //    avoid the expensive JNI call to JVM_Notify() and JVM_NotifyAll().
 //
 // *  use jccb and jmpb instead of jcc and jmp to improve code density.
 //    But beware of excessive branch density on AMD Opterons.
 //
 // *  Both Fast_Lock and Fast_Unlock set the ICC.ZF to indicate success
 //    or failure of the fast-path.  If the fast-path fails then we pass
 //    control to the slow-path, typically in C.  In Fast_Lock and
 //    Fast_Unlock we often branch to DONE_LABEL, just to find that C2
 //    will emit a conditional branch immediately after the node.
 //    So we have branches to branches and lots of ICC.ZF games.
 //    Instead, it might be better to have C2 pass a "FailureLabel"
 //    into Fast_Lock and Fast_Unlock.  In the case of success, control
 //    will drop through the node.  ICC.ZF is undefined at exit.
 //    In the case of failure, the node will branch directly to the
 //    FailureLabel
 // obj: object to lock
 // box: on-stack box address (displaced header location) - KILLED
 // rax,: tmp -- KILLED
 // scr: tmp -- KILLED
 void MacroAssembler::fast_lock(Register objReg, Register boxReg, Register tmpReg, Register scrReg) {
   tmpReg = T8;
   scrReg = S7;
   // Ensure the register assignents are disjoint
   guarantee (objReg != boxReg, "") ;
   guarantee (objReg != tmpReg, "") ;
   guarantee (objReg != scrReg, "") ;
   guarantee (boxReg != tmpReg, "") ;
   guarantee (boxReg != scrReg, "") ;
   block_comment("FastLock");
   /*
      __ move(AT, 0x0);
      return;
      */
   if (PrintBiasedLockingStatistics) {
     push(tmpReg);
     atomic_inc32((address)BiasedLocking::total_entry_count_addr(), 1, AT, tmpReg);
     pop(tmpReg);
   }
   if (EmitSync & 1) {
     // set box->dhw = unused_mark (3)
     // Force all sync thru slow-path: slow_enter() and slow_exit()
     move (AT, (int32_t)intptr_t(markOopDesc::unused_mark()));
     sd(AT, Address(boxReg, 0));
     move (AT, (int32_t)0) ;	// Eflags.ZF = 0
   } else
     if (EmitSync & 2) {
       Label DONE_LABEL ;
       if (UseBiasedLocking) {
         // Note: tmpReg maps to the swap_reg argument and scrReg to the tmp_reg argument.
         biased_locking_enter(boxReg, objReg, tmpReg, scrReg, false, DONE_LABEL, NULL);
       }
       ld(tmpReg, Address(objReg, 0)) ;          // fetch markword
       ori(tmpReg, tmpReg, 0x1);
       sd(tmpReg, Address(boxReg, 0));           // Anticipate successful CAS
       cmpxchg(boxReg, Address(objReg, 0), tmpReg);          // Updates tmpReg
       bne(AT, R0, DONE_LABEL);
       delayed()->nop();
       // Recursive locking
       dsubu(tmpReg, tmpReg, SP);
       li(AT, (7 - os::vm_page_size() ));
       andr(tmpReg, tmpReg, AT);
       sd(tmpReg, Address(boxReg, 0));
       bind(DONE_LABEL) ;
     } else {
       // Possible cases that we'll encounter in fast_lock
       // ------------------------------------------------
       // * Inflated
       //    -- unlocked
       //    -- Locked
       //       = by self
       //       = by other
       // * biased
       //    -- by Self
       //    -- by other
       // * neutral
       // * stack-locked
       //    -- by self
       //       = sp-proximity test hits
       //       = sp-proximity test generates false-negative
       //    -- by other
       //
       Label IsInflated, DONE_LABEL, PopDone ;
       // TODO: optimize away redundant LDs of obj->mark and improve the markword triage
       // order to reduce the number of conditional branches in the most common cases.
       // Beware -- there's a subtle invariant that fetch of the markword
       // at [FETCH], below, will never observe a biased encoding (*101b).
       // If this invariant is not held we risk exclusion (safety) failure.
       if (UseBiasedLocking && !UseOptoBiasInlining) {
         biased_locking_enter(boxReg, objReg, tmpReg, scrReg, false, DONE_LABEL, NULL);
       }
       ld(tmpReg, Address(objReg, 0)) ;         //Fetch the markword of the object.
       andi(AT, tmpReg, 0x02);                  //If AT == 0x02 ==> the object is inflated, will not use the fast lock method.
       bne(AT, R0, IsInflated);                      // Inflated v (Stack-locked or neutral)
       delayed()->nop();
       // Attempt stack-locking ...
       ori (tmpReg, tmpReg, 0x1);
       sd(tmpReg, Address(boxReg, 0));          // Anticipate successful CAS
       cmpxchg(boxReg, Address(objReg, 0), tmpReg);           // Updates tmpReg
       if (PrintBiasedLockingStatistics) {
         Label L;
         beq(AT, R0, L);
         delayed()->nop();
         push(T0);
         push(T1);
         atomic_inc32((address)BiasedLocking::fast_path_entry_count_addr(), 1, T0, T1);
         pop(T1);
         pop(T0);
         bind(L);
       }
       bne(AT, R0, DONE_LABEL);
       delayed()->nop();
       // Recursive locking
       dsubu(tmpReg, tmpReg, SP);
       li(AT, 7 - os::vm_page_size() );
       andr(tmpReg, tmpReg, AT);
       sd(tmpReg, Address(boxReg, 0));
       if (PrintBiasedLockingStatistics) {
         Label L;
         // tmpReg == 0 => BiasedLocking::_fast_path_entry_count++
         bne(tmpReg, R0, L);
         delayed()->nop();
         push(T0);
         push(T1);
         atomic_inc32((address)BiasedLocking::fast_path_entry_count_addr(), 1, T0, T1);
         pop(T1);
         pop(T0);
         bind(L);
       }
       sltiu(AT, tmpReg, 1); /* AT = (tmpReg == 0) ? 1 : 0 */
       b(DONE_LABEL) ;
       delayed()->nop();
       bind(IsInflated) ;
       // TODO: someday avoid the ST-before-CAS penalty by
       // relocating (deferring) the following ST.
       // We should also think about trying a CAS without having
       // fetched _owner.  If the CAS is successful we may
       // avoid an RTO->RTS upgrade on the $line.
       // Without cast to int32_t a movptr will destroy r10 which is typically obj
       li(AT, (int32_t)intptr_t(markOopDesc::unused_mark()));
       sd(AT, Address(boxReg, 0));
       move(boxReg, tmpReg) ;
       ld(tmpReg, Address(tmpReg, ObjectMonitor::owner_offset_in_bytes()-2)) ;
       sltiu(AT, tmpReg, 1);  /* Jin: AT = !tmpReg; */
       bne(tmpReg, R0, DONE_LABEL);
       delayed()->nop();
       cmpxchg(TREG, Address(boxReg, ObjectMonitor::owner_offset_in_bytes()-2), tmpReg) ;
       // Intentional fall-through into DONE_LABEL ...
       // DONE_LABEL is a hot target - we'd really like to place it at the
       // start of cache line by padding with NOPs.
       // See the AMD and Intel software optimization manuals for the
       // most efficient "long" NOP encodings.
       // Unfortunately none of our alignment mechanisms suffice.
       bind(DONE_LABEL);
       // Avoid branch-to-branch on AMD processors
       // This appears to be superstition.
       if (EmitSync & 32) nop() ;
       // At DONE_LABEL the icc ZFlag is set as follows ...
       // Fast_Unlock uses the same protocol.
       // ZFlag == 1 -> Success
       // ZFlag == 0 -> Failure - force control through the slow-path
     }
 }
 // obj: object to unlock
 // box: box address (displaced header location), killed.  Must be EAX.
 // rbx,: killed tmp; cannot be obj nor box.
 //
 // Some commentary on balanced locking:
 //
 // Fast_Lock and Fast_Unlock are emitted only for provably balanced lock sites.
 // Methods that don't have provably balanced locking are forced to run in the
 // interpreter - such methods won't be compiled to use fast_lock and fast_unlock.
 // The interpreter provides two properties:
 // I1:  At return-time the interpreter automatically and quietly unlocks any
 //      objects acquired the current activation (frame).  Recall that the
 //      interpreter maintains an on-stack list of locks currently held by
 //      a frame.
 // I2:  If a method attempts to unlock an object that is not held by the
 //      the frame the interpreter throws IMSX.
 //
 // Lets say A(), which has provably balanced locking, acquires O and then calls B().
 // B() doesn't have provably balanced locking so it runs in the interpreter.
 // Control returns to A() and A() unlocks O.  By I1 and I2, above, we know that O
 // is still locked by A().
 //
 // The only other source of unbalanced locking would be JNI.  The "Java Native Interface:
 // Programmer's Guide and Specification" claims that an object locked by jni_monitorenter
 // should not be unlocked by "normal" java-level locking and vice-versa.  The specification
 // doesn't specify what will occur if a program engages in such mixed-mode locking, however.
 void MacroAssembler::fast_unlock(Register objReg, Register boxReg, Register tmpReg) {
   tmpReg = T8;
   guarantee (objReg != boxReg, "") ;
   guarantee (objReg != tmpReg, "") ;
   guarantee (boxReg != tmpReg, "") ;
   block_comment("FastUnlock");
   /*
      move(AT, 0x0);
      return;
      */
   if (EmitSync & 4) {
     // Disable - inhibit all inlining.  Force control through the slow-path
     move(AT, R0);
   } else
     if (EmitSync & 8) {
       Label DONE_LABEL ;
       if (UseBiasedLocking) {
         biased_locking_exit(objReg, tmpReg, DONE_LABEL);
       }
       // classic stack-locking code ...
       ld(tmpReg, Address(boxReg, 0)) ;
       beq(tmpReg, R0, DONE_LABEL) ;
       move(AT, 0x1);  // delay slot
       cmpxchg(tmpReg, Address(objReg, 0), boxReg);          // Uses EAX which is box
       bind(DONE_LABEL);
     } else {
       Label DONE_LABEL, Stacked, CheckSucc, Inflated ;
       // Critically, the biased locking test must have precedence over
       // and appear before the (box->dhw == 0) recursive stack-lock test.
       if (UseBiasedLocking && !UseOptoBiasInlining) {
         biased_locking_exit(objReg, tmpReg, DONE_LABEL);
       }
       ld(tmpReg, Address(objReg, 0)) ;       // Examine the object's markword
       ld(AT, Address(boxReg, 0)) ;            // Examine the displaced header
       beq(AT, R0, DONE_LABEL) ;      // 0 indicates recursive stack-lock
       //move(AT, 0x1);
       //delayed()->nop();
       delayed()->daddiu(AT, R0, 0x1);
       andi(AT, tmpReg, markOopDesc::monitor_value) ;                     // Inflated?
       beq(AT, R0, Stacked) ;                     // Inflated?
       delayed()->nop();
       bind(Inflated) ;
       // It's inflated.
       // Despite our balanced locking property we still check that m->_owner == Self
       // as java routines or native JNI code called by this thread might
       // have released the lock.
       // Refer to the comments in synchronizer.cpp for how we might encode extra
       // state in _succ so we can avoid fetching EntryList|cxq.
       //
       // I'd like to add more cases in fast_lock() and fast_unlock() --
       // such as recursive enter and exit -- but we have to be wary of
       // I$ bloat, T$ effects and BP$ effects.
       //
       // If there's no contention try a 1-0 exit.  That is, exit without
       // a costly MEMBAR or CAS.  See synchronizer.cpp for details on how
       // we detect and recover from the race that the 1-0 exit admits.
       //
       // Conceptually Fast_Unlock() must execute a STST|LDST "release" barrier
       // before it STs null into _owner, releasing the lock.  Updates
       // to data protected by the critical section must be visible before
       // we drop the lock (and thus before any other thread could acquire
       // the lock and observe the fields protected by the lock).
       // IA32's memory-model is SPO, so STs are ordered with respect to
       // each other and there's no need for an explicit barrier (fence).
       // See also http://gee.cs.oswego.edu/dl/jmm/cookbook.html.
 #ifdef OPT_THREAD
       move(boxReg, TREG);
 #else
       get_thread (boxReg) ;
 #endif
 #ifndef _LP64
       // Note that we could employ various encoding schemes to reduce
       // the number of loads below (currently 4) to just 2 or 3.
       // Refer to the comments in synchronizer.cpp.
       // In practice the chain of fetches doesn't seem to impact performance, however.
       if ((EmitSync & 65536) == 0 && (EmitSync & 256)) {
         // Attempt to reduce branch density - AMD's branch predictor.
         ld(AT, Address (tmpReg, ObjectMonitor::owner_offset_in_bytes()-2)) ;
         xorr(boxReg, boxReg, AT);
         ld(AT, Address (tmpReg, ObjectMonitor::recursions_offset_in_bytes()-2)) ;
         orr(boxReg, boxReg, AT);
         ld(AT, Address (tmpReg, ObjectMonitor::EntryList_offset_in_bytes()-2)) ;
         orr(boxReg, boxReg, AT);
         ld(AT, Address (tmpReg, ObjectMonitor::cxq_offset_in_bytes()-2)) ;
         orr(boxReg, boxReg, AT);
         bne(boxReg, R0, DONE_LABEL);
         move(AT, R0);	/* delay slot */
         sw(R0, Address (tmpReg, ObjectMonitor::owner_offset_in_bytes()-2)) ;
         b(DONE_LABEL);
         move(AT, 0x1);	/* delay slot */
       } else {
         ld(AT, Address (tmpReg, ObjectMonitor::owner_offset_in_bytes()-2)) ;
         xorr(boxReg, boxReg, AT);
         ld(AT, Address (tmpReg, ObjectMonitor::recursions_offset_in_bytes()-2)) ;
         orr(boxReg, boxReg, AT);
         bne(boxReg, R0, DONE_LABEL);
         move(AT, R0);	/* delay slot */
         ld(boxReg, Address (tmpReg, ObjectMonitor::EntryList_offset_in_bytes()-2)) ;
         ld(AT, Address (tmpReg, ObjectMonitor::cxq_offset_in_bytes()-2)) ;
         orr(boxReg, boxReg, AT);
         bne(boxReg, R0, CheckSucc);
         move(AT, R0);	/* delay slot */
         sd(R0, Address (tmpReg, ObjectMonitor::owner_offset_in_bytes()-2)) ;
         b(DONE_LABEL);
         move(AT, 0x1);	/* delay slot */
       }
       // The Following code fragment (EmitSync & 65536) improves the performance of
       // contended applications and contended synchronization microbenchmarks.
       // Unfortunately the emission of the code - even though not executed - causes regressions
       // in scimark and jetstream, evidently because of $ effects.  Replacing the code
       // with an equal number of never-executed NOPs results in the same regression.
       // We leave it off by default.
       if ((EmitSync & 65536) != 0) {
         Label LSuccess, LGoSlowPath ;
         bind(CheckSucc) ;
         // Optional pre-test ... it's safe to elide this
         if ((EmitSync & 16) == 0) {
           ld(AT, Address (tmpReg, ObjectMonitor::succ_offset_in_bytes()-2)) ;
           beq(AT, R0, LGoSlowPath);
           delayed()->nop();
         }
         // We have a classic Dekker-style idiom:
         //    ST m->_owner = 0 ; MEMBAR; LD m->_succ
         // There are a number of ways to implement the barrier:
         // (1) lock:andl &m->_owner, 0
         //     is fast, but mask doesn't currently support the "ANDL M,IMM32" form.
         //     LOCK: ANDL [ebx+Offset(_Owner)-2], 0
         //     Encodes as 81 31 OFF32 IMM32 or 83 63 OFF8 IMM8
         // (2) If supported, an explicit MFENCE is appealing.
         //     In older IA32 processors MFENCE is slower than lock:add or xchg
         //     particularly if the write-buffer is full as might be the case if
         //     if stores closely precede the fence or fence-equivalent instruction.
         //     In more modern implementations MFENCE appears faster, however.
         // (3) In lieu of an explicit fence, use lock:addl to the top-of-stack
         //     The $lines underlying the top-of-stack should be in M-state.
         //     The locked add instruction is serializing, of course.
         // (4) Use xchg, which is serializing
         //     mov boxReg, 0; xchgl boxReg, [tmpReg + Offset(_owner)-2] also works
         // (5) ST m->_owner = 0 and then execute lock:orl &m->_succ, 0.
         //     The integer condition codes will tell us if succ was 0.
         //     Since _succ and _owner should reside in the same $line and
         //     we just stored into _owner, it's likely that the $line
         //     remains in M-state for the lock:orl.
         //
         // We currently use (3), although it's likely that switching to (2)
         // is correct for the future.
         sd(R0, Address (tmpReg, ObjectMonitor::owner_offset_in_bytes()-2)) ;
         // Ratify _succ remains non-null
         ld(AT, Address (tmpReg, ObjectMonitor::succ_offset_in_bytes()-2)) ;
         bne(AT, R0, LSuccess);
         delayed()->nop();		/* delay slot */
         /*
            masm.cmpptr(Address (tmpReg, ObjectMonitor::succ_offset_in_bytes()-2), 0) ;
            masm.jccb  (Assembler::notZero, LSuccess) ;
            */
         move(boxReg, R0) ;                  // box is really EAX
         cmpxchg(SP, Address(tmpReg, ObjectMonitor::owner_offset_in_bytes()-2), boxReg);
         beq(AT, R0, LSuccess);
         delayed()->nop();
         // Since we're low on registers we installed rsp as a placeholding in _owner.
         // Now install Self over rsp.  This is safe as we're transitioning from
         // non-null to non=null
         get_thread (boxReg) ;
         sd(boxReg, Address (tmpReg, ObjectMonitor::owner_offset_in_bytes()-2)) ;
         // Intentional fall-through into LGoSlowPath ...
         bind(LGoSlowPath) ;
         ori(boxReg, boxReg, 1) ;                      // set ICC.ZF=0 to indicate failure
         b(DONE_LABEL) ;
         move(AT, R0) ;	/* delay slot */
         bind(LSuccess) ;
         move(boxReg, R0) ;                 // set ICC.ZF=1 to indicate success
         b(DONE_LABEL) ;
         move(AT, 0x1) ;	/* delay slot */
       }
       bind (Stacked) ;
       // It's not inflated and it's not recursively stack-locked and it's not biased.
       // It must be stack-locked.
       // Try to reset the header to displaced header.
       // The "box" value on the stack is stable, so we can reload
       // and be assured we observe the same value as above.
       ld(tmpReg, Address(boxReg, 0)) ;
       cmpxchg(tmpReg, Address(objReg, 0), boxReg); // Uses EAX which is box
       // Intention fall-thru into DONE_LABEL
       // DONE_LABEL is a hot target - we'd really like to place it at the
       // start of cache line by padding with NOPs.
       // See the AMD and Intel software optimization manuals for the
       // most efficient "long" NOP encodings.
       // Unfortunately none of our alignment mechanisms suffice.
       if ((EmitSync & 65536) == 0) {
         bind (CheckSucc) ;
       }
 #else // _LP64
       // It's inflated
       ld(AT, Address (tmpReg, ObjectMonitor::owner_offset_in_bytes()-2)) ;
       xorr(boxReg, boxReg, AT);
       ld(AT, Address (tmpReg, ObjectMonitor::recursions_offset_in_bytes()-2)) ;
       orr(boxReg, boxReg, AT);
       move(AT, R0);
       bne(boxReg, R0, DONE_LABEL);
       delayed()->nop();
       ld(boxReg, Address (tmpReg, ObjectMonitor::EntryList_offset_in_bytes()-2)) ;
       ld(AT, Address (tmpReg, ObjectMonitor::cxq_offset_in_bytes()-2)) ;
       orr(boxReg, boxReg, AT);
       move(AT, R0);
       bne(boxReg, R0, CheckSucc);
       delayed()->nop();
       sd(R0, Address (tmpReg, ObjectMonitor::owner_offset_in_bytes()-2)) ;
       move(AT, 0x1);
       b(DONE_LABEL);
       delayed()->nop();
       if ((EmitSync & 65536) == 0) {
         Label LSuccess, LGoSlowPath ;
         bind (CheckSucc);
         ld(AT, Address (tmpReg, ObjectMonitor::succ_offset_in_bytes()-2)) ;
         beq(AT, R0, LGoSlowPath);
         delayed()->nop();
         // I'd much rather use lock:andl m->_owner, 0 as it's faster than the
         // the explicit ST;MEMBAR combination, but masm doesn't currently support
         // "ANDQ M,IMM".  Don't use MFENCE here.  lock:add to TOS, xchg, etc
         // are all faster when the write buffer is populated.
         sd(R0, Address (tmpReg, ObjectMonitor::owner_offset_in_bytes()-2)) ;
         if (os::is_MP()) {
           // lock ();
           //addl (Address(rsp, 0), 0); //?
         }
         ld(AT, Address (tmpReg, ObjectMonitor::succ_offset_in_bytes()-2)) ;
         bne(AT, R0, LSuccess);
         delayed()->nop();
         move(boxReg, R0) ;                  // box is really EAX
         //if (os::is_MP()) { lock(); }
         cmpxchg(SP, Address(tmpReg, ObjectMonitor::owner_offset_in_bytes()-2), boxReg);
         beq(AT, R0, LSuccess);
         delayed()->nop();
         // Intentional fall-through into slow-path
         bind  (LGoSlowPath);
         ori(boxReg, boxReg, 1) ;                      // set ICC.ZF=0 to indicate failure
         move(AT, R0);
         b(DONE_LABEL) ;
         delayed()->nop();
         bind  (LSuccess);
         move(boxReg, R0) ;                 // set ICC.ZF=1 to indicate success
         move(AT, 0x1) ;
         b(DONE_LABEL) ;
         delayed()->nop();
       }
       bind  (Stacked);
       ld(tmpReg, Address(boxReg, 0)) ;
       //if (os::is_MP()) { lock(); }
       cmpxchg(tmpReg, Address(objReg, 0), boxReg); // Uses EAX which is box
       if (EmitSync & 65536) {
         bind (CheckSucc);
       }
 #endif
       bind(DONE_LABEL);
       // Avoid branch to branch on AMD processors
       if (EmitSync & 32768) { nop() ; }
     }
 }
 class ControlWord {
 				public:
 								int32_t _value;
   int  rounding_control() const        { return  (_value >> 10) & 3      ; }
   int  precision_control() const       { return  (_value >>  8) & 3      ; }
   bool precision() const               { return ((_value >>  5) & 1) != 0; }
   bool underflow() const               { return ((_value >>  4) & 1) != 0; }
   bool overflow() const                { return ((_value >>  3) & 1) != 0; }
   bool zero_divide() const             { return ((_value >>  2) & 1) != 0; }
   bool denormalized() const            { return ((_value >>  1) & 1) != 0; }
   bool invalid() const                 { return ((_value >>  0) & 1) != 0; }
   void print() const {
     // rounding control
     const char* rc;
     switch (rounding_control()) {
       case 0: rc = "round near"; break;
       case 1: rc = "round down"; break;
       case 2: rc = "round up  "; break;
       case 3: rc = "chop      "; break;
     };
     // precision control
     const char* pc;
     switch (precision_control()) {
       case 0: pc = "24 bits "; break;
       case 1: pc = "reserved"; break;
       case 2: pc = "53 bits "; break;
       case 3: pc = "64 bits "; break;
     };
     // flags
     char f[9];
     f[0] = ' ';
     f[1] = ' ';
     f[2] = (precision   ()) ? 'P' : 'p';
     f[3] = (underflow   ()) ? 'U' : 'u';
     f[4] = (overflow    ()) ? 'O' : 'o';
     f[5] = (zero_divide ()) ? 'Z' : 'z';
     f[6] = (denormalized()) ? 'D' : 'd';
     f[7] = (invalid     ()) ? 'I' : 'i';
     f[8] = '\x0';
     // output
     printf("%04x  masks = %s, %s, %s", _value & 0xFFFF, f, rc, pc);
   }
 };
 class StatusWord {
  public:
   int32_t _value;
   bool busy() const                    { return ((_value >> 15) & 1) != 0; }
   bool C3() const                      { return ((_value >> 14) & 1) != 0; }
   bool C2() const                      { return ((_value >> 10) & 1) != 0; }
   bool C1() const                      { return ((_value >>  9) & 1) != 0; }
   bool C0() const                      { return ((_value >>  8) & 1) != 0; }
   int  top() const                     { return  (_value >> 11) & 7      ; }
   bool error_status() const            { return ((_value >>  7) & 1) != 0; }
   bool stack_fault() const             { return ((_value >>  6) & 1) != 0; }
   bool precision() const               { return ((_value >>  5) & 1) != 0; }
   bool underflow() const               { return ((_value >>  4) & 1) != 0; }
   bool overflow() const                { return ((_value >>  3) & 1) != 0; }
   bool zero_divide() const             { return ((_value >>  2) & 1) != 0; }
   bool denormalized() const            { return ((_value >>  1) & 1) != 0; }
   bool invalid() const                 { return ((_value >>  0) & 1) != 0; }
   void print() const {
     // condition codes
     char c[5];
     c[0] = (C3()) ? '3' : '-';
     c[1] = (C2()) ? '2' : '-';
     c[2] = (C1()) ? '1' : '-';
     c[3] = (C0()) ? '0' : '-';
     c[4] = '\x0';
     // flags
     char f[9];
     f[0] = (error_status()) ? 'E' : '-';
     f[1] = (stack_fault ()) ? 'S' : '-';
     f[2] = (precision   ()) ? 'P' : '-';
     f[3] = (underflow   ()) ? 'U' : '-';
     f[4] = (overflow    ()) ? 'O' : '-';
     f[5] = (zero_divide ()) ? 'Z' : '-';
     f[6] = (denormalized()) ? 'D' : '-';
     f[7] = (invalid     ()) ? 'I' : '-';
     f[8] = '\x0';
     // output
     printf("%04x  flags = %s, cc =  %s, top = %d", _value & 0xFFFF, f, c, top());
   }
 };
 class TagWord {
  public:
   int32_t _value;
   int tag_at(int i) const              { return (_value >> (i*2)) & 3; }
   void print() const {
     printf("%04x", _value & 0xFFFF);
   }
 };
 class FPU_Register {
  public:
   int32_t _m0;
   int32_t _m1;
   int16_t _ex;
   bool is_indefinite() const           {
     return _ex == -1 && _m1 == (int32_t)0xC0000000 && _m0 == 0;
   }
   void print() const {
     char  sign = (_ex < 0) ? '-' : '+';
     const char* kind = (_ex == 0x7FFF || _ex == (int16_t)-1) ? "NaN" : "   ";
     printf("%c%04hx.%08x%08x  %s", sign, _ex, _m1, _m0, kind);
   };
 };
 class FPU_State {
  public:
   enum {
     register_size       = 10,
     number_of_registers =  8,
     register_mask       =  7
   };
   ControlWord  _control_word;
   StatusWord   _status_word;
   TagWord      _tag_word;
   int32_t      _error_offset;
   int32_t      _error_selector;
   int32_t      _data_offset;
   int32_t      _data_selector;
   int8_t       _register[register_size * number_of_registers];
   int tag_for_st(int i) const          { return _tag_word.tag_at((_status_word.top() + i) & register_mask); }
   FPU_Register* st(int i) const        { return (FPU_Register*)&_register[register_size * i]; }
   const char* tag_as_string(int tag) const {
     switch (tag) {
       case 0: return "valid";
       case 1: return "zero";
       case 2: return "special";
       case 3: return "empty";
     }
     ShouldNotReachHere();
     return NULL;
   }
   void print() const {
     // print computation registers
     { int t = _status_word.top();
       for (int i = 0; i < number_of_registers; i++) {
         int j = (i - t) & register_mask;
         printf("%c r%d = ST%d = ", (j == 0 ? '*' : ' '), i, j);
         st(j)->print();
         printf(" %s\n", tag_as_string(_tag_word.tag_at(i)));
       }
     }
     printf("\n");
     // print control registers
     printf("ctrl = "); _control_word.print(); printf("\n");
     printf("stat = "); _status_word .print(); printf("\n");
     printf("tags = "); _tag_word    .print(); printf("\n");
   }
 };
 class Flag_Register {
  public:
   int32_t _value;
   bool overflow() const                { return ((_value >> 11) & 1) != 0; }
   bool direction() const               { return ((_value >> 10) & 1) != 0; }
   bool sign() const                    { return ((_value >>  7) & 1) != 0; }
   bool zero() const                    { return ((_value >>  6) & 1) != 0; }
   bool auxiliary_carry() const         { return ((_value >>  4) & 1) != 0; }
   bool parity() const                  { return ((_value >>  2) & 1) != 0; }
   bool carry() const                   { return ((_value >>  0) & 1) != 0; }
   void print() const {
     // flags
     char f[8];
     f[0] = (overflow       ()) ? 'O' : '-';
     f[1] = (direction      ()) ? 'D' : '-';
     f[2] = (sign           ()) ? 'S' : '-';
     f[3] = (zero           ()) ? 'Z' : '-';
     f[4] = (auxiliary_carry()) ? 'A' : '-';
     f[5] = (parity         ()) ? 'P' : '-';
     f[6] = (carry          ()) ? 'C' : '-';
     f[7] = '\x0';
     // output
     printf("%08x  flags = %s", _value, f);
   }
 };
 class IU_Register {
  public:
   int32_t _value;
   void print() const {
     printf("%08x  %11d", _value, _value);
   }
 };
 class IU_State {
  public:
   Flag_Register _eflags;
   IU_Register   _rdi;
   IU_Register   _rsi;
   IU_Register   _rbp;
   IU_Register   _rsp;
   IU_Register   _rbx;
   IU_Register   _rdx;
   IU_Register   _rcx;
   IU_Register   _rax;
   void print() const {
     // computation registers
     printf("rax,  = "); _rax.print(); printf("\n");
     printf("rbx,  = "); _rbx.print(); printf("\n");
     printf("rcx  = "); _rcx.print(); printf("\n");
     printf("rdx  = "); _rdx.print(); printf("\n");
     printf("rdi  = "); _rdi.print(); printf("\n");
     printf("rsi  = "); _rsi.print(); printf("\n");
     printf("rbp,  = "); _rbp.print(); printf("\n");
     printf("rsp  = "); _rsp.print(); printf("\n");
     printf("\n");
     // control registers
     printf("flgs = "); _eflags.print(); printf("\n");
   }
 };
 class CPU_State {
  public:
   FPU_State _fpu_state;
   IU_State  _iu_state;
   void print() const {
     printf("--------------------------------------------------\n");
     _iu_state .print();
     printf("\n");
     _fpu_state.print();
     printf("--------------------------------------------------\n");
   }
 };
 /*
 static void _print_CPU_state(CPU_State* state) {
   state->print();
 };
 void MacroAssembler::print_CPU_state() {
   push_CPU_state();
   push(rsp);                // pass CPU state
   call(RuntimeAddress(CAST_FROM_FN_PTR(address, _print_CPU_state)));
   addptr(rsp, wordSize);       // discard argument
   pop_CPU_state();
 }
 */
 void MacroAssembler::align(int modulus) {
 	while (offset() % modulus != 0) nop();
 }
 #if 0
 static bool _verify_FPU(int stack_depth, char* s, CPU_State* state) {
   static int counter = 0;
   FPU_State* fs = &state->_fpu_state;
   counter++;
   // For leaf calls, only verify that the top few elements remain empty.
   // We only need 1 empty at the top for C2 code.
   if( stack_depth < 0 ) {
     if( fs->tag_for_st(7) != 3 ) {
       printf("FPR7 not empty\n");
       state->print();
       assert(false, "error");
       return false;
     }
     return true;                // All other stack states do not matter
   }
   assert((fs->_control_word._value & 0xffff) == StubRoutines::_fpu_cntrl_wrd_std,
          "bad FPU control word");
   // compute stack depth
   int i = 0;
   while (i < FPU_State::number_of_registers && fs->tag_for_st(i)  < 3) i++;
   int d = i;
   while (i < FPU_State::number_of_registers && fs->tag_for_st(i) == 3) i++;
   // verify findings
   if (i != FPU_State::number_of_registers) {
     // stack not contiguous
     printf("%s: stack not contiguous at ST%d\n", s, i);
     state->print();
     assert(false, "error");
     return false;
   }
   // check if computed stack depth corresponds to expected stack depth
   if (stack_depth < 0) {
     // expected stack depth is -stack_depth or less
     if (d > -stack_depth) {
       // too many elements on the stack
       printf("%s: <= %d stack elements expected but found %d\n", s, -stack_depth, d);
       state->print();
       assert(false, "error");
       return false;
     }
   } else {
     // expected stack depth is stack_depth
     if (d != stack_depth) {
       // wrong stack depth
       printf("%s: %d stack elements expected but found %d\n", s, stack_depth, d);
       state->print();
       assert(false, "error");
       return false;
     }
   }
   // everything is cool
   return true;
 }
 #endif
 void MacroAssembler::verify_FPU(int stack_depth, const char* s) {
 	//FIXME aoqi
 	// %%%%% need to implement this
 	//Unimplemented();
 	/*
 	if (!VerifyFPU) return;
   push_CPU_state();
   push(rsp);                // pass CPU state
   ExternalAddress msg((address) s);
   // pass message string s
   pushptr(msg.addr());
   push(stack_depth);        // pass stack depth
   call(RuntimeAddress(CAST_FROM_FN_PTR(address, _verify_FPU)));
   addptr(rsp, 3 * wordSize);   // discard arguments
   // check for error
   { Label L;
     testl(rax, rax);
     jcc(Assembler::notZero, L);
     int3();                  // break if error condition
     bind(L);
   }
   pop_CPU_state();
 	*/
 }
 #ifdef _LP64
 Register caller_saved_registers[] = {AT, V0, V1, A0, A1, A2, A3, A4, A5, A6, A7, T0, T1, T2, T3, T8, T9, GP, RA, FP};
 /* FIXME: Jin: In MIPS64, F0~23 are all caller-saved registers */
 FloatRegister caller_saved_fpu_registers[] = {F0, F12, F13};
 #else
 Register caller_saved_registers[] = {AT, V0, V1, A0, A1, A2, A3, T4, T5, T6, T7, T0, T1, T2, T3, T8, T9, GP, RA, FP};
 Register caller_saved_fpu_registers[] = {};
 #endif
 //We preserve all caller-saved register
 void  MacroAssembler::pushad(){
   int i;
   /* Fixed-point registers */
   int len = sizeof(caller_saved_registers) / sizeof(caller_saved_registers[0]);
   daddi(SP, SP, -1 * len * wordSize);
   for (i = 0; i < len; i++)
   {
 #ifdef _LP64
     sd(caller_saved_registers[i], SP, (len - i - 1) * wordSize);
 #else
     sw(caller_saved_registers[i], SP, (len - i - 1) * wordSize);
 #endif
   }
   /* Floating-point registers */
   len = sizeof(caller_saved_fpu_registers) / sizeof(caller_saved_fpu_registers[0]);
   daddi(SP, SP, -1 * len * wordSize);
   for (i = 0; i < len; i++)
   {
 #ifdef _LP64
     sdc1(caller_saved_fpu_registers[i], SP, (len - i - 1) * wordSize);
 #else
     swc1(caller_saved_fpu_registers[i], SP, (len - i - 1) * wordSize);
 #endif
   }
 };
 void  MacroAssembler::popad(){
   int i;
   /* Floating-point registers */
   int len = sizeof(caller_saved_fpu_registers) / sizeof(caller_saved_fpu_registers[0]);
   for (i = 0; i < len; i++)
   {
 #ifdef _LP64
     ldc1(caller_saved_fpu_registers[i], SP, (len - i - 1) * wordSize);
 #else
     lwc1(caller_saved_fpu_registers[i], SP, (len - i - 1) * wordSize);
 #endif
   }
   daddi(SP, SP, len * wordSize);
   /* Fixed-point registers */
   len = sizeof(caller_saved_registers) / sizeof(caller_saved_registers[0]);
   for (i = 0; i < len; i++)
   {
 #ifdef _LP64
     ld(caller_saved_registers[i], SP, (len - i - 1) * wordSize);
 #else
     lw(caller_saved_registers[i], SP, (len - i - 1) * wordSize);
 #endif
   }
   daddi(SP, SP, len * wordSize);
 };
 void MacroAssembler::push2(Register reg1, Register reg2) {
 #ifdef _LP64
   daddi(SP, SP, -16);
   sd(reg2, SP, 0);
   sd(reg1, SP, 8);
 #else
   addi(SP, SP, -8);
   sw(reg2, SP, 0);
   sw(reg1, SP, 4);
 #endif
 }
 void MacroAssembler::pop2(Register reg1, Register reg2) {
 #ifdef _LP64
   ld(reg1, SP, 0);
   ld(reg2, SP, 8);
   daddi(SP, SP, 16);
 #else
   lw(reg1, SP, 0);
   lw(reg2, SP, 4);
   addi(SP, SP, 8);
 #endif
 }
 //for UseCompressedOops Option
 void MacroAssembler::load_klass(Register dst, Register src) {
 #ifdef _LP64
     if(UseCompressedClassPointers){
         lwu(dst, Address(src, oopDesc::klass_offset_in_bytes()));
 		decode_klass_not_null(dst);
     } else
 #endif
         ld(dst, src, oopDesc::klass_offset_in_bytes());
 }
 void MacroAssembler::store_klass(Register dst, Register src) {
 #ifdef _LP64
     if(UseCompressedClassPointers){
 		encode_klass_not_null(src);
 		sw(src, dst, oopDesc::klass_offset_in_bytes());
     } else {
 #endif
 		sd(src, dst, oopDesc::klass_offset_in_bytes());
     }
 }
 void MacroAssembler::load_prototype_header(Register dst, Register src) {
   load_klass(dst, src);
   ld(dst, Address(dst, Klass::prototype_header_offset()));
 }
 #ifdef _LP64
 void MacroAssembler::store_klass_gap(Register dst, Register src) {
   if (UseCompressedClassPointers) {
     sw(src, dst, oopDesc::klass_gap_offset_in_bytes());
   }
 }
 void MacroAssembler::load_heap_oop(Register dst, Address src) {
     if(UseCompressedOops){
 	lwu(dst, src);
 	decode_heap_oop(dst);
     } else{
 	ld(dst, src);
     }
 }
 void MacroAssembler::store_heap_oop(Address dst, Register src){
     if(UseCompressedOops){
        assert(!dst.uses(src), "not enough registers");
        encode_heap_oop(src);
        sw(src, dst);
     } else{
        sd(src, dst);
     }
 }
 #ifdef ASSERT
 void MacroAssembler::verify_heapbase(const char* msg) {
   assert (UseCompressedOops || UseCompressedClassPointers, "should be compressed");
   assert (Universe::heap() != NULL, "java heap should be initialized");
 /*  if (CheckCompressedOops) {
     Label ok;
     push(rscratch1); // cmpptr trashes rscratch1
     cmpptr(r12_heapbase, ExternalAddress((address)Universe::narrow_ptrs_base_addr()));
     jcc(Assembler::equal, ok);
     STOP(msg);
     bind(ok);
     pop(rscratch1);
   }*/
 }
 #endif
 // Algorithm must match oop.inline.hpp encode_heap_oop.
 void MacroAssembler::encode_heap_oop(Register r) {
 #ifdef ASSERT
   verify_heapbase("MacroAssembler::encode_heap_oop:heap base corrupted?");
 #endif
   verify_oop(r, "broken oop in encode_heap_oop");
   if (Universe::narrow_oop_base() == NULL) {
     if (Universe::narrow_oop_shift() != 0) {
       assert (LogMinObjAlignmentInBytes == Universe::narrow_oop_shift(), "decode alg wrong");
       shr(r, LogMinObjAlignmentInBytes);
     }
     return;
   }
     Label done;
     beq(r, R0, done);
     delayed()->nop();
     dsub(r, r, S5_heapbase);
     shr(r, LogMinObjAlignmentInBytes);
     bind(done);
 }
 void MacroAssembler::encode_heap_oop_not_null(Register r) {
     assert (UseCompressedOops, "should be compressed");
 #ifdef ASSERT
     if (CheckCompressedOops) {
 	Label ok;
 	bne(r, R0, ok);
 	delayed()->nop();
 	stop("null oop passed to encode_heap_oop_not_null");
 	bind(ok);
     }
 #endif
 	verify_oop(r, "broken oop in encode_heap_oop_not_null");
 	if (Universe::narrow_oop_base() != NULL) {
 		dsub(r, r, S5_heapbase);
 	}
 	if (Universe::narrow_oop_shift() != 0) {
 		assert (LogMinObjAlignmentInBytes == Universe::narrow_oop_shift(), "decode alg wrong");
 		shr(r, LogMinObjAlignmentInBytes);
 	}
 }
 void MacroAssembler::encode_heap_oop_not_null(Register dst, Register src) {
     assert (UseCompressedOops, "should be compressed");
 #ifdef ASSERT
     if (CheckCompressedOops) {
 	Label ok;
 	bne(src, R0, ok);
 	delayed()->nop();
 	stop("null oop passed to encode_heap_oop_not_null2");
 	bind(ok);
     }
 #endif
     verify_oop(src, "broken oop in encode_heap_oop_not_null2");
     if (dst != src) {
 	move(dst, src);
     }
 	if (Universe::narrow_oop_base() != NULL) {
 		dsub(dst, dst, S5_heapbase);
 	}
 	if (Universe::narrow_oop_shift() != 0) {
 		assert (LogMinObjAlignmentInBytes == Universe::narrow_oop_shift(), "decode alg wrong");
 		shr(dst, LogMinObjAlignmentInBytes);
 	}
 }
 void  MacroAssembler::decode_heap_oop(Register r) {
 #ifdef ASSERT
   verify_heapbase("MacroAssembler::decode_heap_oop corrupted?");
 #endif
   if (Universe::narrow_oop_base() == NULL) {
     if (Universe::narrow_oop_shift() != 0) {
       assert (LogMinObjAlignmentInBytes == Universe::narrow_oop_shift(), "decode alg wrong");
       shl(r, LogMinObjAlignmentInBytes);
     }
   } else {
     Label done;
     shl(r, LogMinObjAlignmentInBytes);
     beq(r, R0, done);
     delayed()->nop();
     dadd(r, r, S5_heapbase);
     bind(done);
   }
   verify_oop(r, "broken oop in decode_heap_oop");
 }
 void  MacroAssembler::decode_heap_oop_not_null(Register r) {
   // Note: it will change flags
   assert (UseCompressedOops, "should only be used for compressed headers");
   assert (Universe::heap() != NULL, "java heap should be initialized");
   // Cannot assert, unverified entry point counts instructions (see .ad file)
   // vtableStubs also counts instructions in pd_code_size_limit.
   // Also do not verify_oop as this is called by verify_oop.
   if (Universe::narrow_oop_shift() != 0) {
     assert(LogMinObjAlignmentInBytes == Universe::narrow_oop_shift(), "decode alg wrong");
     shl(r, LogMinObjAlignmentInBytes);
     if (Universe::narrow_oop_base() != NULL) {
       dadd(r, r, S5_heapbase);
     }
   } else {
     assert (Universe::narrow_oop_base() == NULL, "sanity");
   }
 }
 void  MacroAssembler::decode_heap_oop_not_null(Register dst, Register src) {
   assert (UseCompressedOops, "should only be used for compressed headers");
   assert (Universe::heap() != NULL, "java heap should be initialized");
   // Cannot assert, unverified entry point counts instructions (see .ad file)
   // vtableStubs also counts instructions in pd_code_size_limit.
   // Also do not verify_oop as this is called by verify_oop.
   //lea(dst, Address(S5_heapbase, src, Address::times_8, 0));
   if (Universe::narrow_oop_shift() != 0) {
     assert(LogMinObjAlignmentInBytes == Universe::narrow_oop_shift(), "decode alg wrong");
     if (LogMinObjAlignmentInBytes == Address::times_8) {
       dsll(dst, src, LogMinObjAlignmentInBytes);
       dadd(dst, dst, S5_heapbase);
     } else {
       if (dst != src) {
         move(dst, src);
       }
       shl(dst, LogMinObjAlignmentInBytes);
       if (Universe::narrow_oop_base() != NULL) {
         dadd(dst, dst, S5_heapbase);
       }
     }
   } else {
     assert (Universe::narrow_oop_base() == NULL, "sanity");
     if (dst != src) {
       move(dst, src);
     }
   }
 }
 void MacroAssembler::encode_klass_not_null(Register r) {
   if (Universe::narrow_klass_base() != NULL) {
     // Use r12 as a scratch register in which to temporarily load the narrow_klass_base.
     assert(r != S5_heapbase, "Encoding a klass in r12");
     li48(S5_heapbase, (int64_t)Universe::narrow_klass_base());
     dsub(r, r, S5_heapbase);
   }
   if (Universe::narrow_klass_shift() != 0) {
     assert (LogKlassAlignmentInBytes == Universe::narrow_klass_shift(), "decode alg wrong");
     shr(r, LogKlassAlignmentInBytes);
   }
   if (Universe::narrow_klass_base() != NULL) {
     reinit_heapbase();
   }
 }
 void MacroAssembler::encode_klass_not_null(Register dst, Register src) {
   if (dst == src) {
     encode_klass_not_null(src);
   } else {
     if (Universe::narrow_klass_base() != NULL) {
       li48(dst, (int64_t)Universe::narrow_klass_base());
       dsub(dst, src, dst);
     } else {
       move(dst, src);
     }
     if (Universe::narrow_klass_shift() != 0) {
       assert (LogKlassAlignmentInBytes == Universe::narrow_klass_shift(), "decode alg wrong");
       shr(dst, LogKlassAlignmentInBytes);
     }
   }
 }
 // Function instr_size_for_decode_klass_not_null() counts the instructions
 // generated by decode_klass_not_null(register r) and reinit_heapbase(),
 // when (Universe::heap() != NULL).  Hence, if the instructions they
 // generate change, then this method needs to be updated.
 int MacroAssembler::instr_size_for_decode_klass_not_null() {
   assert (UseCompressedClassPointers, "only for compressed klass ptrs");
   if (Universe::narrow_klass_base() != NULL) {
     // mov64 + addq + shlq? + mov64  (for reinit_heapbase()).
     return (Universe::narrow_klass_shift() == 0 ? 4 * 9 : 4 * 10);
   } else {
     // longest load decode klass function, mov64, leaq
     return (Universe::narrow_klass_shift() == 0 ? 4 * 0 : 4 * 1);
   }
 }
 void  MacroAssembler::decode_klass_not_null(Register r) {
   // Note: it will change flags
   assert (UseCompressedClassPointers, "should only be used for compressed headers");
   assert(r != S5_heapbase, "Decoding a klass in r12");
   // Cannot assert, unverified entry point counts instructions (see .ad file)
   // vtableStubs also counts instructions in pd_code_size_limit.
   // Also do not verify_oop as this is called by verify_oop.
   if (Universe::narrow_klass_shift() != 0) {
     assert(LogKlassAlignmentInBytes == Universe::narrow_klass_shift(), "decode alg wrong");
     shl(r, LogKlassAlignmentInBytes);
   }
   if (Universe::narrow_klass_base() != NULL) {
     li48(S5_heapbase, (int64_t)Universe::narrow_klass_base());
     dadd(r, r, S5_heapbase);
     reinit_heapbase();
   }
 }
 void  MacroAssembler::decode_klass_not_null(Register dst, Register src) {
   assert (UseCompressedClassPointers, "should only be used for compressed headers");
   if (dst == src) {
     decode_klass_not_null(dst);
   } else {
     // Cannot assert, unverified entry point counts instructions (see .ad file)
     // vtableStubs also counts instructions in pd_code_size_limit.
     // Also do not verify_oop as this is called by verify_oop.
     li48(S5_heapbase, (int64_t)Universe::narrow_klass_base());
     if (Universe::narrow_klass_shift() != 0) {
       assert(LogKlassAlignmentInBytes == Universe::narrow_klass_shift(), "decode alg wrong");
       assert(LogKlassAlignmentInBytes == Address::times_8, "klass not aligned on 64bits?");
       dsll(dst, src, Address::times_8);
       dadd(dst, dst, S5_heapbase);
     } else {
       dadd(dst, src, S5_heapbase);
     }
     reinit_heapbase();
   }
 }
 /*
 void  MacroAssembler::set_narrow_oop(Register dst, jobject obj) {
   assert(oop_recorder() != NULL, "this assembler needs an OopRecorder");
   int oop_index = oop_recorder()->find_index(obj);
   RelocationHolder rspec = oop_Relocation::spec(oop_index);
   mov_literal32(dst, oop_index, rspec, narrow_oop_operand);
 }
 */
 void MacroAssembler::incrementl(Register reg, int value) {
   if (value == min_jint) {
      move(AT, value);
      LP64_ONLY(addu32(reg, reg, AT)) NOT_LP64(addu(reg, reg, AT));
      return;
   }
   if (value <  0) { decrementl(reg, -value); return; }
   if (value == 0) {                        ; return; }
   if(Assembler::is_simm16(value)) {
      NOT_LP64(addiu(reg, reg, value));
      LP64_ONLY(move(AT, value); addu32(reg, reg, AT));
   } else {
      move(AT, value);
      LP64_ONLY(addu32(reg, reg, AT)) NOT_LP64(addu(reg, reg, AT));
   }
 }
 void MacroAssembler::decrementl(Register reg, int value) {
   if (value == min_jint) {
      move(AT, value);
      LP64_ONLY(subu32(reg, reg, AT)) NOT_LP64(subu(reg, reg, AT));
      return;
   }
   if (value <  0) { incrementl(reg, -value); return; }
   if (value == 0) {                        ; return; }
   if(Assembler::is_simm16(value)) {
      NOT_LP64(addiu(reg, reg, -value));
      LP64_ONLY(move(AT, value); subu32(reg, reg, AT));
   } else {
      move(AT, value);
      LP64_ONLY(subu32(reg, reg, AT)) NOT_LP64(subu(reg, reg, AT));
   }
 }
 void MacroAssembler::reinit_heapbase() {
   if (UseCompressedOops || UseCompressedClassPointers) {
     if (Universe::heap() != NULL) {
       if (Universe::narrow_oop_base() == NULL) {
         move(S5_heapbase, R0);
       } else {
         li48(S5_heapbase, (int64_t)Universe::narrow_ptrs_base());
       }
     } else {
       li48(S5_heapbase, (intptr_t)Universe::narrow_ptrs_base_addr());
       ld(S5_heapbase, S5_heapbase, 0);
     }
   }
 }
 #endif // _LP64
 void MacroAssembler::check_klass_subtype(Register sub_klass,
                            Register super_klass,
                            Register temp_reg,
                            Label& L_success) {
 //implement ind   gen_subtype_check
   Label L_failure;
   check_klass_subtype_fast_path(sub_klass, super_klass, temp_reg,        &L_success, &L_failure, NULL);
   check_klass_subtype_slow_path(sub_klass, super_klass, temp_reg, noreg, &L_success, NULL);
   bind(L_failure);
 }
 SkipIfEqual::SkipIfEqual(
     MacroAssembler* masm, const bool* flag_addr, bool value) {
   _masm = masm;
   _masm->li(AT, (address)flag_addr);
   _masm->lb(AT,AT,0);
   _masm->addi(AT,AT,-value);
   _masm->beq(AT,R0,_label);
   _masm->delayed()->nop();
 }
 void MacroAssembler::check_klass_subtype_fast_path(Register sub_klass,
                                                    Register super_klass,
                                                    Register temp_reg,
                                                    Label* L_success,
                                                    Label* L_failure,
                                                    Label* L_slow_path,
                                         RegisterOrConstant super_check_offset) {
   assert_different_registers(sub_klass, super_klass, temp_reg);
   bool must_load_sco = (super_check_offset.constant_or_zero() == -1);
   if (super_check_offset.is_register()) {
     assert_different_registers(sub_klass, super_klass,
                                super_check_offset.as_register());
   } else if (must_load_sco) {
     assert(temp_reg != noreg, "supply either a temp or a register offset");
   }
   Label L_fallthrough;
   int label_nulls = 0;
   if (L_success == NULL)   { L_success   = &L_fallthrough; label_nulls++; }
   if (L_failure == NULL)   { L_failure   = &L_fallthrough; label_nulls++; }
   if (L_slow_path == NULL) { L_slow_path = &L_fallthrough; label_nulls++; }
   assert(label_nulls <= 1, "at most one NULL in the batch");
   int sc_offset = in_bytes(Klass::secondary_super_cache_offset());
   int sco_offset = in_bytes(Klass::super_check_offset_offset());
   // If the pointers are equal, we are done (e.g., String[] elements).
   // This self-check enables sharing of secondary supertype arrays among
   // non-primary types such as array-of-interface.  Otherwise, each such
   // type would need its own customized SSA.
   // We move this check to the front of the fast path because many
   // type checks are in fact trivially successful in this manner,
   // so we get a nicely predicted branch right at the start of the check.
   //cmpptr(sub_klass, super_klass);
   //local_jcc(Assembler::equal, *L_success);
   beq(sub_klass, super_klass, *L_success);
   delayed()->nop();
   // Check the supertype display:
   if (must_load_sco) {
     // Positive movl does right thing on LP64.
 	lwu(temp_reg, super_klass, sco_offset);
     super_check_offset = RegisterOrConstant(temp_reg);
   }
   dsll(AT, super_check_offset.register_or_noreg(), Address::times_1);
   daddu(AT, sub_klass, AT);
   ld(AT, AT, super_check_offset.constant_or_zero()*Address::times_1);
   // This check has worked decisively for primary supers.
   // Secondary supers are sought in the super_cache ('super_cache_addr').
   // (Secondary supers are interfaces and very deeply nested subtypes.)
   // This works in the same check above because of a tricky aliasing
   // between the super_cache and the primary super display elements.
   // (The 'super_check_addr' can address either, as the case requires.)
   // Note that the cache is updated below if it does not help us find
   // what we need immediately.
   // So if it was a primary super, we can just fail immediately.
   // Otherwise, it's the slow path for us (no success at this point).
   if (super_check_offset.is_register()) {
 	beq(super_klass, AT, *L_success);
 	delayed()->nop();
 	addi(AT, super_check_offset.as_register(), -sc_offset);
     if (L_failure == &L_fallthrough) {
 	  beq(AT, R0, *L_slow_path);
 	  delayed()->nop();
     } else {
 	  bne(AT, R0, *L_failure);
 	  delayed()->nop();
 	  b(*L_slow_path);
 	  delayed()->nop();
     }
   } else if (super_check_offset.as_constant() == sc_offset) {
     // Need a slow path; fast failure is impossible.
     if (L_slow_path == &L_fallthrough) {
 		beq(super_klass, AT, *L_success);
 		delayed()->nop();
     } else {
 		bne(super_klass, AT, *L_slow_path);
 		delayed()->nop();
 		b(*L_success);
 		delayed()->nop();
     }
   } else {
     // No slow path; it's a fast decision.
     if (L_failure == &L_fallthrough) {
 		beq(super_klass, AT, *L_success);
 		delayed()->nop();
     } else {
 		bne(super_klass, AT, *L_failure);
 		delayed()->nop();
 		b(*L_success);
 		delayed()->nop();
     }
   }
   bind(L_fallthrough);
 }
 void MacroAssembler::check_klass_subtype_slow_path(Register sub_klass,
                                                    Register super_klass,
                                                    Register temp_reg,
                                                    Register temp2_reg,
                                                    Label* L_success,
                                                    Label* L_failure,
                                                    bool set_cond_codes) {
   assert_different_registers(sub_klass, super_klass, temp_reg);
   if (temp2_reg != noreg)
     assert_different_registers(sub_klass, super_klass, temp_reg, temp2_reg);
   else
     temp2_reg = T9;
 #define IS_A_TEMP(reg) ((reg) == temp_reg || (reg) == temp2_reg)
   Label L_fallthrough;
   int label_nulls = 0;
   if (L_success == NULL)   { L_success   = &L_fallthrough; label_nulls++; }
   if (L_failure == NULL)   { L_failure   = &L_fallthrough; label_nulls++; }
   assert(label_nulls <= 1, "at most one NULL in the batch");
   // a couple of useful fields in sub_klass:
   int ss_offset = in_bytes(Klass::secondary_supers_offset());
   int sc_offset = in_bytes(Klass::secondary_super_cache_offset());
   Address secondary_supers_addr(sub_klass, ss_offset);
   Address super_cache_addr(     sub_klass, sc_offset);
   // Do a linear scan of the secondary super-klass chain.
   // This code is rarely used, so simplicity is a virtue here.
   // The repne_scan instruction uses fixed registers, which we must spill.
   // Don't worry too much about pre-existing connections with the input regs.
 #if 0
   assert(sub_klass != T9, "killed reg"); // killed by mov(rax, super)
   assert(sub_klass != T1, "killed reg"); // killed by lea(rcx, &pst_counter)
 #endif
   // Get super_klass value into rax (even if it was in rdi or rcx).
 /*
   bool pushed_rax = false, pushed_rcx = false, pushed_rdi = false;
   if (super_klass != rax || UseCompressedOops) {
     if (!IS_A_TEMP(rax)) { push(rax); pushed_rax = true; }
     mov(rax, super_klass);
   }
   if (!IS_A_TEMP(rcx)) { push(rcx); pushed_rcx = true; }
   if (!IS_A_TEMP(rdi)) { push(rdi); pushed_rdi = true; }
 */
 #ifndef PRODUCT
   int* pst_counter = &SharedRuntime::_partial_subtype_ctr;
   ExternalAddress pst_counter_addr((address) pst_counter);
   NOT_LP64(  incrementl(pst_counter_addr) );
   //LP64_ONLY( lea(rcx, pst_counter_addr) );
   //LP64_ONLY( incrementl(Address(rcx, 0)) );
 #endif //PRODUCT
   // We will consult the secondary-super array.
   ld(temp_reg, secondary_supers_addr);
   // Load the array length.  (Positive movl does right thing on LP64.)
   lw(temp2_reg, Address(temp_reg, Array<Klass*>::length_offset_in_bytes()));
   // Skip to start of data.
   daddiu(temp_reg, temp_reg, Array<Klass*>::base_offset_in_bytes());
   // Scan RCX words at [RDI] for an occurrence of RAX.
   // Set NZ/Z based on last compare.
   // Z flag value will not be set by 'repne' if RCX == 0 since 'repne' does
   // not change flags (only scas instruction which is repeated sets flags).
   // Set Z = 0 (not equal) before 'repne' to indicate that class was not found.
   /* 2013/4/3 Jin: OpenJDK8 never compresses klass pointers in secondary-super array. */
   Label Loop, subtype;
   bind(Loop);
   beq(temp2_reg, R0, *L_failure);
   delayed()->nop();
   ld(AT, temp_reg, 0);
   beq(AT, super_klass, subtype);
   delayed()->daddi(temp_reg, temp_reg, 1 * wordSize);
   b(Loop);
   delayed()->daddi(temp2_reg, temp2_reg, -1);
   bind(subtype);
   sd(super_klass, super_cache_addr);
   if (L_success != &L_fallthrough) {
 	  b(*L_success);
 	  delayed()->nop();
   }
 /*
   if (set_cond_codes) {
     // Special hack for the AD files:  rdi is guaranteed non-zero.
     assert(!pushed_rdi, "rdi must be left non-NULL");
     // Also, the condition codes are properly set Z/NZ on succeed/failure.
   }
 */
   // Success.  Cache the super we found and proceed in triumph.
 #undef IS_A_TEMP
   bind(L_fallthrough);
 }
 void MacroAssembler::get_vm_result(Register oop_result, Register java_thread) {
   ld(oop_result, Address(java_thread, JavaThread::vm_result_offset()));
   sd(R0, Address(java_thread, JavaThread::vm_result_offset()));
   verify_oop(oop_result, "broken oop in call_VM_base");
 }
 void MacroAssembler::get_vm_result_2(Register metadata_result, Register java_thread) {
   ld(metadata_result, Address(java_thread, JavaThread::vm_result_2_offset()));
   sd(R0, Address(java_thread, JavaThread::vm_result_2_offset()));
 }
 Address MacroAssembler::argument_address(RegisterOrConstant arg_slot,
                                          int extra_slot_offset) {
   // cf. TemplateTable::prepare_invoke(), if (load_receiver).
   int stackElementSize = Interpreter::stackElementSize;
   int offset = Interpreter::expr_offset_in_bytes(extra_slot_offset+0);
 #ifdef ASSERT
   int offset1 = Interpreter::expr_offset_in_bytes(extra_slot_offset+1);
   assert(offset1 - offset == stackElementSize, "correct arithmetic");
 #endif
   Register             scale_reg    = NOREG;
   Address::ScaleFactor scale_factor = Address::no_scale;
   if (arg_slot.is_constant()) {
     offset += arg_slot.as_constant() * stackElementSize;
   } else {
     scale_reg    = arg_slot.as_register();
     scale_factor = Address::times_8;
   }
   // 2014/07/31 Fu: We don't push RA on stack in prepare_invoke.
   //  offset += wordSize;           // return PC is on stack
   if(scale_reg==NOREG) return Address(SP, offset);
   else {
 	dsll(scale_reg, scale_reg, scale_factor);
 	daddu(scale_reg, SP, scale_reg);
 	return Address(scale_reg, offset);
   }
 }
 SkipIfEqual::~SkipIfEqual() {
   _masm->bind(_label);
 }
 void MacroAssembler::load_sized_value(Register dst, Address src, size_t size_in_bytes, bool is_signed, Register dst2) {
   switch (size_in_bytes) {
 #ifndef _LP64
   case  8:
     assert(dst2 != noreg, "second dest register required");
     lw(dst,  src);
     lw(dst2, src.plus_disp(BytesPerInt));
     break;
 #else
   case  8:  ld(dst, src); break;
 #endif
   case  4:  lw(dst, src); break;
   case  2:  is_signed ? lh(dst, src) : lhu(dst, src); break;
   case  1:  is_signed ? lb( dst, src) : lbu( dst, src); break;
   default:  ShouldNotReachHere();
   }
 }
 void MacroAssembler::store_sized_value(Address dst, Register src, size_t size_in_bytes, Register src2) {
   switch (size_in_bytes) {
 #ifndef _LP64
   case  8:
     assert(src2 != noreg, "second source register required");
     sw(src, dst);
     sw(src2, dst.plus_disp(BytesPerInt));
     break;
 #else
   case  8:  sd(src, dst); break;
 #endif
   case  4:  sw(src, dst); break;
   case  2:  sh(src, dst); break;
   case  1:  sb(src, dst); break;
   default:  ShouldNotReachHere();
   }
 }
 // Look up the method for a megamorphic invokeinterface call.
 // The target method is determined by <intf_klass, itable_index>.
 // The receiver klass is in recv_klass.
 // On success, the result will be in method_result, and execution falls through.
 // On failure, execution transfers to the given label.
 void MacroAssembler::lookup_interface_method(Register recv_klass,
                                              Register intf_klass,
                                              RegisterOrConstant itable_index,
                                              Register method_result,
                                              Register scan_temp,
                                              Label& L_no_such_interface) {
   assert_different_registers(recv_klass, intf_klass, method_result, scan_temp);
   assert(itable_index.is_constant() || itable_index.as_register() == method_result,
          "caller must use same register for non-constant itable index as for method");
   // Compute start of first itableOffsetEntry (which is at the end of the vtable)
   int vtable_base = InstanceKlass::vtable_start_offset() * wordSize;
   int itentry_off = itableMethodEntry::method_offset_in_bytes();
   int scan_step   = itableOffsetEntry::size() * wordSize;
   int vte_size    = vtableEntry::size() * wordSize;
   Address::ScaleFactor times_vte_scale = Address::times_ptr;
   assert(vte_size == wordSize, "else adjust times_vte_scale");
   lw(scan_temp, Address(recv_klass, InstanceKlass::vtable_length_offset() * wordSize));
   // %%% Could store the aligned, prescaled offset in the klassoop.
 //  lea(scan_temp, Address(recv_klass, scan_temp, times_vte_scale, vtable_base));
   dsll(scan_temp, scan_temp, times_vte_scale);
   daddu(scan_temp, recv_klass, scan_temp);
   daddiu(scan_temp, scan_temp, vtable_base);
   if (HeapWordsPerLong > 1) {
     // Round up to align_object_offset boundary
     // see code for InstanceKlass::start_of_itable!
     round_to(scan_temp, BytesPerLong);
   }
   // Adjust recv_klass by scaled itable_index, so we can free itable_index.
   assert(itableMethodEntry::size() * wordSize == wordSize, "adjust the scaling in the code below");
 //  lea(recv_klass, Address(recv_klass, itable_index, Address::times_ptr, itentry_off));
   if (itable_index.is_constant()) {
     li48(AT, (int)itable_index.is_constant());
     dsll(AT, AT, (int)Address::times_ptr);
   } else {
     dsll(AT, itable_index.as_register(), (int)Address::times_ptr);
   }
   daddu(AT, AT, recv_klass);
   daddiu(recv_klass, AT, itentry_off);
   // for (scan = klass->itable(); scan->interface() != NULL; scan += scan_step) {
   //   if (scan->interface() == intf) {
   //     result = (klass + scan->offset() + itable_index);
   //   }
   // }
   Label search, found_method;
   for (int peel = 1; peel >= 0; peel--) {
     ld(method_result, Address(scan_temp, itableOffsetEntry::interface_offset_in_bytes()));
     if (peel) {
       beq(intf_klass, method_result, found_method);
       nop();
     } else {
       bne(intf_klass, method_result, search);
       nop();
       // (invert the test to fall through to found_method...)
     }
     if (!peel)  break;
     bind(search);
     // Check that the previous entry is non-null.  A null entry means that
     // the receiver class doesn't implement the interface, and wasn't the
     // same as when the caller was compiled.
     beq(method_result, R0, L_no_such_interface);
     nop();
     daddiu(scan_temp, scan_temp, scan_step);
   }
   bind(found_method);
   // Got a hit.
   lw(scan_temp, Address(scan_temp, itableOffsetEntry::offset_offset_in_bytes()));
   ld(method_result, Address(recv_klass, scan_temp, Address::times_1));
 }
 // virtual method calling
 void MacroAssembler::lookup_virtual_method(Register recv_klass,
                                            RegisterOrConstant vtable_index,
                                            Register method_result) {
   Register tmp = GP;
   push(tmp);
   if (vtable_index.is_constant()) {
     assert_different_registers(recv_klass, method_result, tmp);
   } else {
     assert_different_registers(recv_klass, method_result, vtable_index.as_register(), tmp);
   }
   const int base = InstanceKlass::vtable_start_offset() * wordSize;
   assert(vtableEntry::size() * wordSize == wordSize, "else adjust the scaling in the code below");
 /*
   Address vtable_entry_addr(recv_klass,
                             vtable_index, Address::times_ptr,
                             base + vtableEntry::method_offset_in_bytes());
 */
   if (vtable_index.is_constant()) {
     li48(AT, vtable_index.as_constant());
     dsll(AT, AT, (int)Address::times_ptr);
   } else {
     dsll(AT, vtable_index.as_register(), (int)Address::times_ptr);
   }
   li48(tmp, base + vtableEntry::method_offset_in_bytes());
   daddu(tmp, tmp, AT);
   daddu(tmp, tmp, recv_klass);
   ld(method_result, tmp, 0);
   pop(tmp);
 }

Mercurial > jdk8-mips64-public > hotspot / annotate

src/cpu/mips/vm/assembler_mips.cpp@45b68a4063c2 (annotated)

src/cpu/mips/vm/assembler_mips.cpp