8076373: In 32-bit VM interpreter and compiled code process NaN values differently
Summary: Change interpreter to use XMM registers on x86_32 if they are available. Add stubs for methods transforming from/to int/long float/double.
Reviewed-by: kvn, mcberg
/*
* Copyright (c) 1997, 2015, Oracle and/or its affiliates. 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/macroAssembler.hpp"
#include "asm/macroAssembler.inline.hpp"
#include "compiler/disassembler.hpp"
#include "interpreter/bytecodeHistogram.hpp"
#include "interpreter/bytecodeInterpreter.hpp"
#include "interpreter/interpreter.hpp"
#include "interpreter/interpreterGenerator.hpp"
#include "interpreter/interpreterRuntime.hpp"
#include "interpreter/interp_masm.hpp"
#include "interpreter/templateTable.hpp"
#include "memory/allocation.inline.hpp"
#include "memory/resourceArea.hpp"
#include "oops/arrayOop.hpp"
#include "oops/methodData.hpp"
#include "oops/method.hpp"
#include "oops/oop.inline.hpp"
#include "prims/forte.hpp"
#include "prims/jvmtiExport.hpp"
#include "prims/methodHandles.hpp"
#include "runtime/handles.inline.hpp"
#include "runtime/sharedRuntime.hpp"
#include "runtime/stubRoutines.hpp"
#include "runtime/timer.hpp"
# define __ _masm->
//------------------------------------------------------------------------------------------------------------------------
// Implementation of InterpreterCodelet
void InterpreterCodelet::initialize(const char* description, Bytecodes::Code bytecode) {
_description = description;
_bytecode = bytecode;
}
void InterpreterCodelet::verify() {
}
void InterpreterCodelet::print_on(outputStream* st) const {
ttyLocker ttyl;
if (PrintInterpreter) {
st->cr();
st->print_cr("----------------------------------------------------------------------");
}
if (description() != NULL) st->print("%s ", description());
if (bytecode() >= 0 ) st->print("%d %s ", bytecode(), Bytecodes::name(bytecode()));
st->print_cr("[" INTPTR_FORMAT ", " INTPTR_FORMAT "] %d bytes",
p2i(code_begin()), p2i(code_end()), code_size());
if (PrintInterpreter) {
st->cr();
Disassembler::decode(code_begin(), code_end(), st, DEBUG_ONLY(_strings) NOT_DEBUG(CodeStrings()));
}
}
CodeletMark::CodeletMark(InterpreterMacroAssembler*& masm,
const char* description,
Bytecodes::Code bytecode) :
_clet((InterpreterCodelet*)AbstractInterpreter::code()->request(codelet_size())),
_cb(_clet->code_begin(), _clet->code_size()) {
// Request all space (add some slack for Codelet data).
assert(_clet != NULL, "we checked not enough space already");
// Initialize Codelet attributes.
_clet->initialize(description, bytecode);
// Create assembler for code generation.
masm = new InterpreterMacroAssembler(&_cb);
_masm = &masm;
}
CodeletMark::~CodeletMark() {
// Align so printing shows nop's instead of random code at the end (Codelets are aligned).
(*_masm)->align(wordSize);
// Make sure all code is in code buffer.
(*_masm)->flush();
// Commit Codelet.
AbstractInterpreter::code()->commit((*_masm)->code()->pure_insts_size(), (*_masm)->code()->strings());
// Make sure nobody can use _masm outside a CodeletMark lifespan.
*_masm = NULL;
}
//------------------------------------------------------------------------------------------------------------------------
// Implementation of platform independent aspects of Interpreter
void AbstractInterpreter::initialize() {
if (_code != NULL) return;
// make sure 'imported' classes are initialized
if (CountBytecodes || TraceBytecodes || StopInterpreterAt) BytecodeCounter::reset();
if (PrintBytecodeHistogram) BytecodeHistogram::reset();
if (PrintBytecodePairHistogram) BytecodePairHistogram::reset();
InvocationCounter::reinitialize(DelayCompilationDuringStartup);
}
void AbstractInterpreter::print() {
tty->cr();
tty->print_cr("----------------------------------------------------------------------");
tty->print_cr("Interpreter");
tty->cr();
tty->print_cr("code size = %6dK bytes", (int)_code->used_space()/1024);
tty->print_cr("total space = %6dK bytes", (int)_code->total_space()/1024);
tty->print_cr("wasted space = %6dK bytes", (int)_code->available_space()/1024);
tty->cr();
tty->print_cr("# of codelets = %6d" , _code->number_of_stubs());
if (_code->number_of_stubs() != 0) {
tty->print_cr("avg codelet size = %6d bytes", _code->used_space() / _code->number_of_stubs());
tty->cr();
}
_code->print();
tty->print_cr("----------------------------------------------------------------------");
tty->cr();
}
void interpreter_init() {
Interpreter::initialize();
#ifndef PRODUCT
if (TraceBytecodes) BytecodeTracer::set_closure(BytecodeTracer::std_closure());
#endif // PRODUCT
// need to hit every safepoint in order to call zapping routine
// register the interpreter
Forte::register_stub(
"Interpreter",
AbstractInterpreter::code()->code_start(),
AbstractInterpreter::code()->code_end()
);
// notify JVMTI profiler
if (JvmtiExport::should_post_dynamic_code_generated()) {
JvmtiExport::post_dynamic_code_generated("Interpreter",
AbstractInterpreter::code()->code_start(),
AbstractInterpreter::code()->code_end());
}
}
//------------------------------------------------------------------------------------------------------------------------
// Implementation of interpreter
StubQueue* AbstractInterpreter::_code = NULL;
bool AbstractInterpreter::_notice_safepoints = false;
address AbstractInterpreter::_rethrow_exception_entry = NULL;
address AbstractInterpreter::_native_entry_begin = NULL;
address AbstractInterpreter::_native_entry_end = NULL;
address AbstractInterpreter::_slow_signature_handler;
address AbstractInterpreter::_entry_table [AbstractInterpreter::number_of_method_entries];
address AbstractInterpreter::_native_abi_to_tosca [AbstractInterpreter::number_of_result_handlers];
//------------------------------------------------------------------------------------------------------------------------
// Generation of complete interpreter
AbstractInterpreterGenerator::AbstractInterpreterGenerator(StubQueue* _code) {
_masm = NULL;
}
static const BasicType types[Interpreter::number_of_result_handlers] = {
T_BOOLEAN,
T_CHAR ,
T_BYTE ,
T_SHORT ,
T_INT ,
T_LONG ,
T_VOID ,
T_FLOAT ,
T_DOUBLE ,
T_OBJECT
};
void AbstractInterpreterGenerator::generate_all() {
{ CodeletMark cm(_masm, "slow signature handler");
Interpreter::_slow_signature_handler = generate_slow_signature_handler();
}
}
//------------------------------------------------------------------------------------------------------------------------
// Entry points
AbstractInterpreter::MethodKind AbstractInterpreter::method_kind(methodHandle m) {
// Abstract method?
if (m->is_abstract()) return abstract;
// Method handle primitive?
if (m->is_method_handle_intrinsic()) {
vmIntrinsics::ID id = m->intrinsic_id();
assert(MethodHandles::is_signature_polymorphic(id), "must match an intrinsic");
MethodKind kind = (MethodKind)( method_handle_invoke_FIRST +
((int)id - vmIntrinsics::FIRST_MH_SIG_POLY) );
assert(kind <= method_handle_invoke_LAST, "parallel enum ranges");
return kind;
}
#ifndef CC_INTERP
if (UseCRC32Intrinsics && m->is_native()) {
// Use optimized stub code for CRC32 native methods.
switch (m->intrinsic_id()) {
case vmIntrinsics::_updateCRC32 : return java_util_zip_CRC32_update;
case vmIntrinsics::_updateBytesCRC32 : return java_util_zip_CRC32_updateBytes;
case vmIntrinsics::_updateByteBufferCRC32 : return java_util_zip_CRC32_updateByteBuffer;
}
}
switch(m->intrinsic_id()) {
case vmIntrinsics::_intBitsToFloat: return java_lang_Float_intBitsToFloat;
case vmIntrinsics::_floatToRawIntBits: return java_lang_Float_floatToRawIntBits;
case vmIntrinsics::_longBitsToDouble: return java_lang_Double_longBitsToDouble;
case vmIntrinsics::_doubleToRawLongBits: return java_lang_Double_doubleToRawLongBits;
}
#endif // CC_INTERP
// Native method?
// Note: This test must come _before_ the test for intrinsic
// methods. See also comments below.
if (m->is_native()) {
assert(!m->is_method_handle_intrinsic(), "overlapping bits here, watch out");
return m->is_synchronized() ? native_synchronized : native;
}
// Synchronized?
if (m->is_synchronized()) {
return zerolocals_synchronized;
}
if (RegisterFinalizersAtInit && m->code_size() == 1 &&
m->intrinsic_id() == vmIntrinsics::_Object_init) {
// We need to execute the special return bytecode to check for
// finalizer registration so create a normal frame.
return zerolocals;
}
// Empty method?
if (m->is_empty_method()) {
return empty;
}
// Special intrinsic method?
// Note: This test must come _after_ the test for native methods,
// otherwise we will run into problems with JDK 1.2, see also
// InterpreterGenerator::generate_method_entry() for
// for details.
switch (m->intrinsic_id()) {
case vmIntrinsics::_dsin : return java_lang_math_sin ;
case vmIntrinsics::_dcos : return java_lang_math_cos ;
case vmIntrinsics::_dtan : return java_lang_math_tan ;
case vmIntrinsics::_dabs : return java_lang_math_abs ;
case vmIntrinsics::_dsqrt : return java_lang_math_sqrt ;
case vmIntrinsics::_dlog : return java_lang_math_log ;
case vmIntrinsics::_dlog10: return java_lang_math_log10;
case vmIntrinsics::_dpow : return java_lang_math_pow ;
case vmIntrinsics::_dexp : return java_lang_math_exp ;
case vmIntrinsics::_Reference_get:
return java_lang_ref_reference_get;
}
// Accessor method?
if (m->is_accessor()) {
assert(m->size_of_parameters() == 1, "fast code for accessors assumes parameter size = 1");
return accessor;
}
// Note: for now: zero locals for all non-empty methods
return zerolocals;
}
void AbstractInterpreter::set_entry_for_kind(AbstractInterpreter::MethodKind kind, address entry) {
assert(kind >= method_handle_invoke_FIRST &&
kind <= method_handle_invoke_LAST, "late initialization only for MH entry points");
assert(_entry_table[kind] == _entry_table[abstract], "previous value must be AME entry");
_entry_table[kind] = entry;
}
// Return true if the interpreter can prove that the given bytecode has
// not yet been executed (in Java semantics, not in actual operation).
bool AbstractInterpreter::is_not_reached(methodHandle method, int bci) {
Bytecodes::Code code = method()->code_at(bci);
if (!Bytecodes::must_rewrite(code)) {
// might have been reached
return false;
}
// the bytecode might not be rewritten if the method is an accessor, etc.
address ientry = method->interpreter_entry();
if (ientry != entry_for_kind(AbstractInterpreter::zerolocals) &&
ientry != entry_for_kind(AbstractInterpreter::zerolocals_synchronized))
return false; // interpreter does not run this method!
// otherwise, we can be sure this bytecode has never been executed
return true;
}
#ifndef PRODUCT
void AbstractInterpreter::print_method_kind(MethodKind kind) {
switch (kind) {
case zerolocals : tty->print("zerolocals" ); break;
case zerolocals_synchronized: tty->print("zerolocals_synchronized"); break;
case native : tty->print("native" ); break;
case native_synchronized : tty->print("native_synchronized" ); break;
case empty : tty->print("empty" ); break;
case accessor : tty->print("accessor" ); break;
case abstract : tty->print("abstract" ); break;
case java_lang_math_sin : tty->print("java_lang_math_sin" ); break;
case java_lang_math_cos : tty->print("java_lang_math_cos" ); break;
case java_lang_math_tan : tty->print("java_lang_math_tan" ); break;
case java_lang_math_abs : tty->print("java_lang_math_abs" ); break;
case java_lang_math_sqrt : tty->print("java_lang_math_sqrt" ); break;
case java_lang_math_log : tty->print("java_lang_math_log" ); break;
case java_lang_math_log10 : tty->print("java_lang_math_log10" ); break;
case java_util_zip_CRC32_update : tty->print("java_util_zip_CRC32_update"); break;
case java_util_zip_CRC32_updateBytes : tty->print("java_util_zip_CRC32_updateBytes"); break;
case java_util_zip_CRC32_updateByteBuffer : tty->print("java_util_zip_CRC32_updateByteBuffer"); break;
default:
if (kind >= method_handle_invoke_FIRST &&
kind <= method_handle_invoke_LAST) {
const char* kind_name = vmIntrinsics::name_at(method_handle_intrinsic(kind));
if (kind_name[0] == '_') kind_name = &kind_name[1]; // '_invokeExact' => 'invokeExact'
tty->print("method_handle_%s", kind_name);
break;
}
ShouldNotReachHere();
break;
}
}
#endif // PRODUCT
//------------------------------------------------------------------------------------------------------------------------
// Deoptimization support
/**
* If a deoptimization happens, this function returns the point of next bytecode to continue execution.
*/
address AbstractInterpreter::deopt_continue_after_entry(Method* method, address bcp, int callee_parameters, bool is_top_frame) {
assert(method->contains(bcp), "just checkin'");
// Get the original and rewritten bytecode.
Bytecodes::Code code = Bytecodes::java_code_at(method, bcp);
assert(!Interpreter::bytecode_should_reexecute(code), "should not reexecute");
const int bci = method->bci_from(bcp);
// compute continuation length
const int length = Bytecodes::length_at(method, bcp);
// compute result type
BasicType type = T_ILLEGAL;
switch (code) {
case Bytecodes::_invokevirtual :
case Bytecodes::_invokespecial :
case Bytecodes::_invokestatic :
case Bytecodes::_invokeinterface: {
Thread *thread = Thread::current();
ResourceMark rm(thread);
methodHandle mh(thread, method);
type = Bytecode_invoke(mh, bci).result_type();
// since the cache entry might not be initialized:
// (NOT needed for the old calling convension)
if (!is_top_frame) {
int index = Bytes::get_native_u2(bcp+1);
method->constants()->cache()->entry_at(index)->set_parameter_size(callee_parameters);
}
break;
}
case Bytecodes::_invokedynamic: {
Thread *thread = Thread::current();
ResourceMark rm(thread);
methodHandle mh(thread, method);
type = Bytecode_invoke(mh, bci).result_type();
// since the cache entry might not be initialized:
// (NOT needed for the old calling convension)
if (!is_top_frame) {
int index = Bytes::get_native_u4(bcp+1);
method->constants()->invokedynamic_cp_cache_entry_at(index)->set_parameter_size(callee_parameters);
}
break;
}
case Bytecodes::_ldc :
case Bytecodes::_ldc_w : // fall through
case Bytecodes::_ldc2_w:
{
Thread *thread = Thread::current();
ResourceMark rm(thread);
methodHandle mh(thread, method);
type = Bytecode_loadconstant(mh, bci).result_type();
break;
}
default:
type = Bytecodes::result_type(code);
break;
}
// return entry point for computed continuation state & bytecode length
return
is_top_frame
? Interpreter::deopt_entry (as_TosState(type), length)
: Interpreter::return_entry(as_TosState(type), length, code);
}
// If deoptimization happens, this function returns the point where the interpreter reexecutes
// the bytecode.
// Note: Bytecodes::_athrow is a special case in that it does not return
// Interpreter::deopt_entry(vtos, 0) like others
address AbstractInterpreter::deopt_reexecute_entry(Method* method, address bcp) {
assert(method->contains(bcp), "just checkin'");
Bytecodes::Code code = Bytecodes::java_code_at(method, bcp);
#ifdef COMPILER1
if(code == Bytecodes::_athrow ) {
return Interpreter::rethrow_exception_entry();
}
#endif /* COMPILER1 */
return Interpreter::deopt_entry(vtos, 0);
}
// If deoptimization happens, the interpreter should reexecute these bytecodes.
// This function mainly helps the compilers to set up the reexecute bit.
bool AbstractInterpreter::bytecode_should_reexecute(Bytecodes::Code code) {
switch (code) {
case Bytecodes::_lookupswitch:
case Bytecodes::_tableswitch:
case Bytecodes::_fast_binaryswitch:
case Bytecodes::_fast_linearswitch:
// recompute condtional expression folded into _if<cond>
case Bytecodes::_lcmp :
case Bytecodes::_fcmpl :
case Bytecodes::_fcmpg :
case Bytecodes::_dcmpl :
case Bytecodes::_dcmpg :
case Bytecodes::_ifnull :
case Bytecodes::_ifnonnull :
case Bytecodes::_goto :
case Bytecodes::_goto_w :
case Bytecodes::_ifeq :
case Bytecodes::_ifne :
case Bytecodes::_iflt :
case Bytecodes::_ifge :
case Bytecodes::_ifgt :
case Bytecodes::_ifle :
case Bytecodes::_if_icmpeq :
case Bytecodes::_if_icmpne :
case Bytecodes::_if_icmplt :
case Bytecodes::_if_icmpge :
case Bytecodes::_if_icmpgt :
case Bytecodes::_if_icmple :
case Bytecodes::_if_acmpeq :
case Bytecodes::_if_acmpne :
// special cases
case Bytecodes::_getfield :
case Bytecodes::_putfield :
case Bytecodes::_getstatic :
case Bytecodes::_putstatic :
case Bytecodes::_aastore :
#ifdef COMPILER1
//special case of reexecution
case Bytecodes::_athrow :
#endif
return true;
default:
return false;
}
}
void AbstractInterpreterGenerator::bang_stack_shadow_pages(bool native_call) {
// Quick & dirty stack overflow checking: bang the stack & handle trap.
// Note that we do the banging after the frame is setup, since the exception
// handling code expects to find a valid interpreter frame on the stack.
// Doing the banging earlier fails if the caller frame is not an interpreter
// frame.
// (Also, the exception throwing code expects to unlock any synchronized
// method receiever, so do the banging after locking the receiver.)
// Bang each page in the shadow zone. We can't assume it's been done for
// an interpreter frame with greater than a page of locals, so each page
// needs to be checked. Only true for non-native.
if (UseStackBanging) {
const int start_page = native_call ? StackShadowPages : 1;
const int page_size = os::vm_page_size();
for (int pages = start_page; pages <= StackShadowPages ; pages++) {
__ bang_stack_with_offset(pages*page_size);
}
}
}
void AbstractInterpreterGenerator::initialize_method_handle_entries() {
// method handle entry kinds are generated later in MethodHandlesAdapterGenerator::generate:
for (int i = Interpreter::method_handle_invoke_FIRST; i <= Interpreter::method_handle_invoke_LAST; i++) {
Interpreter::MethodKind kind = (Interpreter::MethodKind) i;
Interpreter::_entry_table[kind] = Interpreter::_entry_table[Interpreter::abstract];
}
}
// Generate method entries
address InterpreterGenerator::generate_method_entry(
AbstractInterpreter::MethodKind kind) {
// determine code generation flags
bool synchronized = false;
address entry_point = NULL;
switch (kind) {
case Interpreter::zerolocals : break;
case Interpreter::zerolocals_synchronized: synchronized = true; break;
case Interpreter::native : entry_point = generate_native_entry(false); break;
case Interpreter::native_synchronized : entry_point = generate_native_entry(true); break;
case Interpreter::empty : entry_point = generate_empty_entry(); break;
case Interpreter::accessor : entry_point = generate_accessor_entry(); break;
case Interpreter::abstract : entry_point = generate_abstract_entry(); break;
case Interpreter::java_lang_math_sin : // fall thru
case Interpreter::java_lang_math_cos : // fall thru
case Interpreter::java_lang_math_tan : // fall thru
case Interpreter::java_lang_math_abs : // fall thru
case Interpreter::java_lang_math_log : // fall thru
case Interpreter::java_lang_math_log10 : // fall thru
case Interpreter::java_lang_math_sqrt : // fall thru
case Interpreter::java_lang_math_pow : // fall thru
case Interpreter::java_lang_math_exp : entry_point = generate_math_entry(kind); break;
case Interpreter::java_lang_ref_reference_get
: entry_point = generate_Reference_get_entry(); break;
#ifndef CC_INTERP
case Interpreter::java_util_zip_CRC32_update
: entry_point = generate_CRC32_update_entry(); break;
case Interpreter::java_util_zip_CRC32_updateBytes
: // fall thru
case Interpreter::java_util_zip_CRC32_updateByteBuffer
: entry_point = generate_CRC32_updateBytes_entry(kind); break;
#if defined(TARGET_ARCH_x86) && !defined(_LP64)
// On x86_32 platforms, a special entry is generated for the following four methods.
// On other platforms the normal entry is used to enter these methods.
case Interpreter::java_lang_Float_intBitsToFloat
: entry_point = generate_Float_intBitsToFloat_entry(); break;
case Interpreter::java_lang_Float_floatToRawIntBits
: entry_point = generate_Float_floatToRawIntBits_entry(); break;
case Interpreter::java_lang_Double_longBitsToDouble
: entry_point = generate_Double_longBitsToDouble_entry(); break;
case Interpreter::java_lang_Double_doubleToRawLongBits
: entry_point = generate_Double_doubleToRawLongBits_entry(); break;
#else
case Interpreter::java_lang_Float_intBitsToFloat:
case Interpreter::java_lang_Float_floatToRawIntBits:
case Interpreter::java_lang_Double_longBitsToDouble:
case Interpreter::java_lang_Double_doubleToRawLongBits:
entry_point = generate_native_entry(false);
break;
#endif // defined(TARGET_ARCH_x86) && !defined(_LP64)
#endif // CC_INTERP
default:
fatal(err_msg("unexpected method kind: %d", kind));
break;
}
if (entry_point) {
return entry_point;
}
return generate_normal_entry(synchronized);
}