//
// Copyright (c) 2011, 2016, Oracle and/or its affiliates. All rights reserved.
// Copyright (c) 2012, 2015 SAP SE. All rights reserved.
// DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
//
// This code is free software; you can redistribute it and/or modify it
// 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.
//
//
//
// PPC64 Architecture Description File
//
//----------REGISTER DEFINITION BLOCK------------------------------------------
// This information is used by the matcher and the register allocator to
// describe individual registers and classes of registers within the target
// architecture.
register %{
//----------Architecture Description Register Definitions----------------------
// General Registers
// "reg_def" name (register save type, C convention save type,
// ideal register type, encoding);
//
// Register Save Types:
//
// NS = No-Save: The register allocator assumes that these registers
// can be used without saving upon entry to the method, &
// that they do not need to be saved at call sites.
//
// SOC = Save-On-Call: The register allocator assumes that these registers
// can be used without saving upon entry to the method,
// but that they must be saved at call sites.
// These are called "volatiles" on ppc.
//
// SOE = Save-On-Entry: The register allocator assumes that these registers
// must be saved before using them upon entry to the
// method, but they do not need to be saved at call
// sites.
// These are called "nonvolatiles" on ppc.
//
// AS = Always-Save: The register allocator assumes that these registers
// must be saved before using them upon entry to the
// method, & that they must be saved at call sites.
//
// Ideal Register Type is used to determine how to save & restore a
// register. Op_RegI will get spilled with LoadI/StoreI, Op_RegP will get
// spilled with LoadP/StoreP. If the register supports both, use Op_RegI.
//
// The encoding number is the actual bit-pattern placed into the opcodes.
//
// PPC64 register definitions, based on the 64-bit PowerPC ELF ABI
// Supplement Version 1.7 as of 2003-10-29.
//
// For each 64-bit register we must define two registers: the register
// itself, e.g. R3, and a corresponding virtual other (32-bit-)'half',
// e.g. R3_H, which is needed by the allocator, but is not used
// for stores, loads, etc.
// ----------------------------
// Integer/Long Registers
// ----------------------------
// PPC64 has 32 64-bit integer registers.
// types: v = volatile, nv = non-volatile, s = system
reg_def R0 ( SOC, SOC, Op_RegI, 0, R0->as_VMReg() ); // v used in prologs
reg_def R0_H ( SOC, SOC, Op_RegI, 99, R0->as_VMReg()->next() );
reg_def R1 ( NS, NS, Op_RegI, 1, R1->as_VMReg() ); // s SP
reg_def R1_H ( NS, NS, Op_RegI, 99, R1->as_VMReg()->next() );
reg_def R2 ( SOC, SOC, Op_RegI, 2, R2->as_VMReg() ); // v TOC
reg_def R2_H ( SOC, SOC, Op_RegI, 99, R2->as_VMReg()->next() );
reg_def R3 ( SOC, SOC, Op_RegI, 3, R3->as_VMReg() ); // v iarg1 & iret
reg_def R3_H ( SOC, SOC, Op_RegI, 99, R3->as_VMReg()->next() );
reg_def R4 ( SOC, SOC, Op_RegI, 4, R4->as_VMReg() ); // iarg2
reg_def R4_H ( SOC, SOC, Op_RegI, 99, R4->as_VMReg()->next() );
reg_def R5 ( SOC, SOC, Op_RegI, 5, R5->as_VMReg() ); // v iarg3
reg_def R5_H ( SOC, SOC, Op_RegI, 99, R5->as_VMReg()->next() );
reg_def R6 ( SOC, SOC, Op_RegI, 6, R6->as_VMReg() ); // v iarg4
reg_def R6_H ( SOC, SOC, Op_RegI, 99, R6->as_VMReg()->next() );
reg_def R7 ( SOC, SOC, Op_RegI, 7, R7->as_VMReg() ); // v iarg5
reg_def R7_H ( SOC, SOC, Op_RegI, 99, R7->as_VMReg()->next() );
reg_def R8 ( SOC, SOC, Op_RegI, 8, R8->as_VMReg() ); // v iarg6
reg_def R8_H ( SOC, SOC, Op_RegI, 99, R8->as_VMReg()->next() );
reg_def R9 ( SOC, SOC, Op_RegI, 9, R9->as_VMReg() ); // v iarg7
reg_def R9_H ( SOC, SOC, Op_RegI, 99, R9->as_VMReg()->next() );
reg_def R10 ( SOC, SOC, Op_RegI, 10, R10->as_VMReg() ); // v iarg8
reg_def R10_H( SOC, SOC, Op_RegI, 99, R10->as_VMReg()->next());
reg_def R11 ( SOC, SOC, Op_RegI, 11, R11->as_VMReg() ); // v ENV / scratch
reg_def R11_H( SOC, SOC, Op_RegI, 99, R11->as_VMReg()->next());
reg_def R12 ( SOC, SOC, Op_RegI, 12, R12->as_VMReg() ); // v scratch
reg_def R12_H( SOC, SOC, Op_RegI, 99, R12->as_VMReg()->next());
reg_def R13 ( NS, NS, Op_RegI, 13, R13->as_VMReg() ); // s system thread id
reg_def R13_H( NS, NS, Op_RegI, 99, R13->as_VMReg()->next());
reg_def R14 ( SOC, SOE, Op_RegI, 14, R14->as_VMReg() ); // nv
reg_def R14_H( SOC, SOE, Op_RegI, 99, R14->as_VMReg()->next());
reg_def R15 ( SOC, SOE, Op_RegI, 15, R15->as_VMReg() ); // nv
reg_def R15_H( SOC, SOE, Op_RegI, 99, R15->as_VMReg()->next());
reg_def R16 ( SOC, SOE, Op_RegI, 16, R16->as_VMReg() ); // nv
reg_def R16_H( SOC, SOE, Op_RegI, 99, R16->as_VMReg()->next());
reg_def R17 ( SOC, SOE, Op_RegI, 17, R17->as_VMReg() ); // nv
reg_def R17_H( SOC, SOE, Op_RegI, 99, R17->as_VMReg()->next());
reg_def R18 ( SOC, SOE, Op_RegI, 18, R18->as_VMReg() ); // nv
reg_def R18_H( SOC, SOE, Op_RegI, 99, R18->as_VMReg()->next());
reg_def R19 ( SOC, SOE, Op_RegI, 19, R19->as_VMReg() ); // nv
reg_def R19_H( SOC, SOE, Op_RegI, 99, R19->as_VMReg()->next());
reg_def R20 ( SOC, SOE, Op_RegI, 20, R20->as_VMReg() ); // nv
reg_def R20_H( SOC, SOE, Op_RegI, 99, R20->as_VMReg()->next());
reg_def R21 ( SOC, SOE, Op_RegI, 21, R21->as_VMReg() ); // nv
reg_def R21_H( SOC, SOE, Op_RegI, 99, R21->as_VMReg()->next());
reg_def R22 ( SOC, SOE, Op_RegI, 22, R22->as_VMReg() ); // nv
reg_def R22_H( SOC, SOE, Op_RegI, 99, R22->as_VMReg()->next());
reg_def R23 ( SOC, SOE, Op_RegI, 23, R23->as_VMReg() ); // nv
reg_def R23_H( SOC, SOE, Op_RegI, 99, R23->as_VMReg()->next());
reg_def R24 ( SOC, SOE, Op_RegI, 24, R24->as_VMReg() ); // nv
reg_def R24_H( SOC, SOE, Op_RegI, 99, R24->as_VMReg()->next());
reg_def R25 ( SOC, SOE, Op_RegI, 25, R25->as_VMReg() ); // nv
reg_def R25_H( SOC, SOE, Op_RegI, 99, R25->as_VMReg()->next());
reg_def R26 ( SOC, SOE, Op_RegI, 26, R26->as_VMReg() ); // nv
reg_def R26_H( SOC, SOE, Op_RegI, 99, R26->as_VMReg()->next());
reg_def R27 ( SOC, SOE, Op_RegI, 27, R27->as_VMReg() ); // nv
reg_def R27_H( SOC, SOE, Op_RegI, 99, R27->as_VMReg()->next());
reg_def R28 ( SOC, SOE, Op_RegI, 28, R28->as_VMReg() ); // nv
reg_def R28_H( SOC, SOE, Op_RegI, 99, R28->as_VMReg()->next());
reg_def R29 ( SOC, SOE, Op_RegI, 29, R29->as_VMReg() ); // nv
reg_def R29_H( SOC, SOE, Op_RegI, 99, R29->as_VMReg()->next());
reg_def R30 ( SOC, SOE, Op_RegI, 30, R30->as_VMReg() ); // nv
reg_def R30_H( SOC, SOE, Op_RegI, 99, R30->as_VMReg()->next());
reg_def R31 ( SOC, SOE, Op_RegI, 31, R31->as_VMReg() ); // nv
reg_def R31_H( SOC, SOE, Op_RegI, 99, R31->as_VMReg()->next());
// ----------------------------
// Float/Double Registers
// ----------------------------
// Double Registers
// The rules of ADL require that double registers be defined in pairs.
// Each pair must be two 32-bit values, but not necessarily a pair of
// single float registers. In each pair, ADLC-assigned register numbers
// must be adjacent, with the lower number even. Finally, when the
// CPU stores such a register pair to memory, the word associated with
// the lower ADLC-assigned number must be stored to the lower address.
// PPC64 has 32 64-bit floating-point registers. Each can store a single
// or double precision floating-point value.
// types: v = volatile, nv = non-volatile, s = system
reg_def F0 ( SOC, SOC, Op_RegF, 0, F0->as_VMReg() ); // v scratch
reg_def F0_H ( SOC, SOC, Op_RegF, 99, F0->as_VMReg()->next() );
reg_def F1 ( SOC, SOC, Op_RegF, 1, F1->as_VMReg() ); // v farg1 & fret
reg_def F1_H ( SOC, SOC, Op_RegF, 99, F1->as_VMReg()->next() );
reg_def F2 ( SOC, SOC, Op_RegF, 2, F2->as_VMReg() ); // v farg2
reg_def F2_H ( SOC, SOC, Op_RegF, 99, F2->as_VMReg()->next() );
reg_def F3 ( SOC, SOC, Op_RegF, 3, F3->as_VMReg() ); // v farg3
reg_def F3_H ( SOC, SOC, Op_RegF, 99, F3->as_VMReg()->next() );
reg_def F4 ( SOC, SOC, Op_RegF, 4, F4->as_VMReg() ); // v farg4
reg_def F4_H ( SOC, SOC, Op_RegF, 99, F4->as_VMReg()->next() );
reg_def F5 ( SOC, SOC, Op_RegF, 5, F5->as_VMReg() ); // v farg5
reg_def F5_H ( SOC, SOC, Op_RegF, 99, F5->as_VMReg()->next() );
reg_def F6 ( SOC, SOC, Op_RegF, 6, F6->as_VMReg() ); // v farg6
reg_def F6_H ( SOC, SOC, Op_RegF, 99, F6->as_VMReg()->next() );
reg_def F7 ( SOC, SOC, Op_RegF, 7, F7->as_VMReg() ); // v farg7
reg_def F7_H ( SOC, SOC, Op_RegF, 99, F7->as_VMReg()->next() );
reg_def F8 ( SOC, SOC, Op_RegF, 8, F8->as_VMReg() ); // v farg8
reg_def F8_H ( SOC, SOC, Op_RegF, 99, F8->as_VMReg()->next() );
reg_def F9 ( SOC, SOC, Op_RegF, 9, F9->as_VMReg() ); // v farg9
reg_def F9_H ( SOC, SOC, Op_RegF, 99, F9->as_VMReg()->next() );
reg_def F10 ( SOC, SOC, Op_RegF, 10, F10->as_VMReg() ); // v farg10
reg_def F10_H( SOC, SOC, Op_RegF, 99, F10->as_VMReg()->next());
reg_def F11 ( SOC, SOC, Op_RegF, 11, F11->as_VMReg() ); // v farg11
reg_def F11_H( SOC, SOC, Op_RegF, 99, F11->as_VMReg()->next());
reg_def F12 ( SOC, SOC, Op_RegF, 12, F12->as_VMReg() ); // v farg12
reg_def F12_H( SOC, SOC, Op_RegF, 99, F12->as_VMReg()->next());
reg_def F13 ( SOC, SOC, Op_RegF, 13, F13->as_VMReg() ); // v farg13
reg_def F13_H( SOC, SOC, Op_RegF, 99, F13->as_VMReg()->next());
reg_def F14 ( SOC, SOE, Op_RegF, 14, F14->as_VMReg() ); // nv
reg_def F14_H( SOC, SOE, Op_RegF, 99, F14->as_VMReg()->next());
reg_def F15 ( SOC, SOE, Op_RegF, 15, F15->as_VMReg() ); // nv
reg_def F15_H( SOC, SOE, Op_RegF, 99, F15->as_VMReg()->next());
reg_def F16 ( SOC, SOE, Op_RegF, 16, F16->as_VMReg() ); // nv
reg_def F16_H( SOC, SOE, Op_RegF, 99, F16->as_VMReg()->next());
reg_def F17 ( SOC, SOE, Op_RegF, 17, F17->as_VMReg() ); // nv
reg_def F17_H( SOC, SOE, Op_RegF, 99, F17->as_VMReg()->next());
reg_def F18 ( SOC, SOE, Op_RegF, 18, F18->as_VMReg() ); // nv
reg_def F18_H( SOC, SOE, Op_RegF, 99, F18->as_VMReg()->next());
reg_def F19 ( SOC, SOE, Op_RegF, 19, F19->as_VMReg() ); // nv
reg_def F19_H( SOC, SOE, Op_RegF, 99, F19->as_VMReg()->next());
reg_def F20 ( SOC, SOE, Op_RegF, 20, F20->as_VMReg() ); // nv
reg_def F20_H( SOC, SOE, Op_RegF, 99, F20->as_VMReg()->next());
reg_def F21 ( SOC, SOE, Op_RegF, 21, F21->as_VMReg() ); // nv
reg_def F21_H( SOC, SOE, Op_RegF, 99, F21->as_VMReg()->next());
reg_def F22 ( SOC, SOE, Op_RegF, 22, F22->as_VMReg() ); // nv
reg_def F22_H( SOC, SOE, Op_RegF, 99, F22->as_VMReg()->next());
reg_def F23 ( SOC, SOE, Op_RegF, 23, F23->as_VMReg() ); // nv
reg_def F23_H( SOC, SOE, Op_RegF, 99, F23->as_VMReg()->next());
reg_def F24 ( SOC, SOE, Op_RegF, 24, F24->as_VMReg() ); // nv
reg_def F24_H( SOC, SOE, Op_RegF, 99, F24->as_VMReg()->next());
reg_def F25 ( SOC, SOE, Op_RegF, 25, F25->as_VMReg() ); // nv
reg_def F25_H( SOC, SOE, Op_RegF, 99, F25->as_VMReg()->next());
reg_def F26 ( SOC, SOE, Op_RegF, 26, F26->as_VMReg() ); // nv
reg_def F26_H( SOC, SOE, Op_RegF, 99, F26->as_VMReg()->next());
reg_def F27 ( SOC, SOE, Op_RegF, 27, F27->as_VMReg() ); // nv
reg_def F27_H( SOC, SOE, Op_RegF, 99, F27->as_VMReg()->next());
reg_def F28 ( SOC, SOE, Op_RegF, 28, F28->as_VMReg() ); // nv
reg_def F28_H( SOC, SOE, Op_RegF, 99, F28->as_VMReg()->next());
reg_def F29 ( SOC, SOE, Op_RegF, 29, F29->as_VMReg() ); // nv
reg_def F29_H( SOC, SOE, Op_RegF, 99, F29->as_VMReg()->next());
reg_def F30 ( SOC, SOE, Op_RegF, 30, F30->as_VMReg() ); // nv
reg_def F30_H( SOC, SOE, Op_RegF, 99, F30->as_VMReg()->next());
reg_def F31 ( SOC, SOE, Op_RegF, 31, F31->as_VMReg() ); // nv
reg_def F31_H( SOC, SOE, Op_RegF, 99, F31->as_VMReg()->next());
// ----------------------------
// Special Registers
// ----------------------------
// Condition Codes Flag Registers
// PPC64 has 8 condition code "registers" which are all contained
// in the CR register.
// types: v = volatile, nv = non-volatile, s = system
reg_def CCR0(SOC, SOC, Op_RegFlags, 0, CCR0->as_VMReg()); // v
reg_def CCR1(SOC, SOC, Op_RegFlags, 1, CCR1->as_VMReg()); // v
reg_def CCR2(SOC, SOC, Op_RegFlags, 2, CCR2->as_VMReg()); // nv
reg_def CCR3(SOC, SOC, Op_RegFlags, 3, CCR3->as_VMReg()); // nv
reg_def CCR4(SOC, SOC, Op_RegFlags, 4, CCR4->as_VMReg()); // nv
reg_def CCR5(SOC, SOC, Op_RegFlags, 5, CCR5->as_VMReg()); // v
reg_def CCR6(SOC, SOC, Op_RegFlags, 6, CCR6->as_VMReg()); // v
reg_def CCR7(SOC, SOC, Op_RegFlags, 7, CCR7->as_VMReg()); // v
// Special registers of PPC64
reg_def SR_XER( SOC, SOC, Op_RegP, 0, SR_XER->as_VMReg()); // v
reg_def SR_LR( SOC, SOC, Op_RegP, 1, SR_LR->as_VMReg()); // v
reg_def SR_CTR( SOC, SOC, Op_RegP, 2, SR_CTR->as_VMReg()); // v
reg_def SR_VRSAVE( SOC, SOC, Op_RegP, 3, SR_VRSAVE->as_VMReg()); // v
reg_def SR_SPEFSCR(SOC, SOC, Op_RegP, 4, SR_SPEFSCR->as_VMReg()); // v
reg_def SR_PPR( SOC, SOC, Op_RegP, 5, SR_PPR->as_VMReg()); // v
// ----------------------------
// Specify priority of register selection within phases of register
// allocation. Highest priority is first. A useful heuristic is to
// give registers a low priority when they are required by machine
// instructions, like EAX and EDX on I486, and choose no-save registers
// before save-on-call, & save-on-call before save-on-entry. Registers
// which participate in fixed calling sequences should come last.
// Registers which are used as pairs must fall on an even boundary.
// It's worth about 1% on SPEC geomean to get this right.
// Chunk0, chunk1, and chunk2 form the MachRegisterNumbers enumeration
// in adGlobals_ppc.hpp which defines the <register>_num values, e.g.
// R3_num. Therefore, R3_num may not be (and in reality is not)
// the same as R3->encoding()! Furthermore, we cannot make any
// assumptions on ordering, e.g. R3_num may be less than R2_num.
// Additionally, the function
// static enum RC rc_class(OptoReg::Name reg )
// maps a given <register>_num value to its chunk type (except for flags)
// and its current implementation relies on chunk0 and chunk1 having a
// size of 64 each.
// If you change this allocation class, please have a look at the
// default values for the parameters RoundRobinIntegerRegIntervalStart
// and RoundRobinFloatRegIntervalStart
alloc_class chunk0 (
// Chunk0 contains *all* 64 integer registers halves.
// "non-volatile" registers
R14, R14_H,
R15, R15_H,
R17, R17_H,
R18, R18_H,
R19, R19_H,
R20, R20_H,
R21, R21_H,
R22, R22_H,
R23, R23_H,
R24, R24_H,
R25, R25_H,
R26, R26_H,
R27, R27_H,
R28, R28_H,
R29, R29_H,
R30, R30_H,
R31, R31_H,
// scratch/special registers
R11, R11_H,
R12, R12_H,
// argument registers
R10, R10_H,
R9, R9_H,
R8, R8_H,
R7, R7_H,
R6, R6_H,
R5, R5_H,
R4, R4_H,
R3, R3_H,
// special registers, not available for allocation
R16, R16_H, // R16_thread
R13, R13_H, // system thread id
R2, R2_H, // may be used for TOC
R1, R1_H, // SP
R0, R0_H // R0 (scratch)
);
// If you change this allocation class, please have a look at the
// default values for the parameters RoundRobinIntegerRegIntervalStart
// and RoundRobinFloatRegIntervalStart
alloc_class chunk1 (
// Chunk1 contains *all* 64 floating-point registers halves.
// scratch register
F0, F0_H,
// argument registers
F13, F13_H,
F12, F12_H,
F11, F11_H,
F10, F10_H,
F9, F9_H,
F8, F8_H,
F7, F7_H,
F6, F6_H,
F5, F5_H,
F4, F4_H,
F3, F3_H,
F2, F2_H,
F1, F1_H,
// non-volatile registers
F14, F14_H,
F15, F15_H,
F16, F16_H,
F17, F17_H,
F18, F18_H,
F19, F19_H,
F20, F20_H,
F21, F21_H,
F22, F22_H,
F23, F23_H,
F24, F24_H,
F25, F25_H,
F26, F26_H,
F27, F27_H,
F28, F28_H,
F29, F29_H,
F30, F30_H,
F31, F31_H
);
alloc_class chunk2 (
// Chunk2 contains *all* 8 condition code registers.
CCR0,
CCR1,
CCR2,
CCR3,
CCR4,
CCR5,
CCR6,
CCR7
);
alloc_class chunk3 (
// special registers
// These registers are not allocated, but used for nodes generated by postalloc expand.
SR_XER,
SR_LR,
SR_CTR,
SR_VRSAVE,
SR_SPEFSCR,
SR_PPR
);
//-------Architecture Description Register Classes-----------------------
// Several register classes are automatically defined based upon
// information in this architecture description.
// 1) reg_class inline_cache_reg ( as defined in frame section )
// 2) reg_class compiler_method_oop_reg ( as defined in frame section )
// 2) reg_class interpreter_method_oop_reg ( as defined in frame section )
// 3) reg_class stack_slots( /* one chunk of stack-based "registers" */ )
//
// ----------------------------
// 32 Bit Register Classes
// ----------------------------
// We specify registers twice, once as read/write, and once read-only.
// We use the read-only registers for source operands. With this, we
// can include preset read only registers in this class, as a hard-coded
// '0'-register. (We used to simulate this on ppc.)
// 32 bit registers that can be read and written i.e. these registers
// can be dest (or src) of normal instructions.
reg_class bits32_reg_rw(
/*R0*/ // R0
/*R1*/ // SP
R2, // TOC
R3,
R4,
R5,
R6,
R7,
R8,
R9,
R10,
R11,
R12,
/*R13*/ // system thread id
R14,
R15,
/*R16*/ // R16_thread
R17,
R18,
R19,
R20,
R21,
R22,
R23,
R24,
R25,
R26,
R27,
R28,
/*R29,*/ // global TOC
R30,
R31
);
// 32 bit registers that can only be read i.e. these registers can
// only be src of all instructions.
reg_class bits32_reg_ro(
/*R0*/ // R0
/*R1*/ // SP
R2 // TOC
R3,
R4,
R5,
R6,
R7,
R8,
R9,
R10,
R11,
R12,
/*R13*/ // system thread id
R14,
R15,
/*R16*/ // R16_thread
R17,
R18,
R19,
R20,
R21,
R22,
R23,
R24,
R25,
R26,
R27,
R28,
/*R29,*/
R30,
R31
);
reg_class rscratch1_bits32_reg(R11);
reg_class rscratch2_bits32_reg(R12);
reg_class rarg1_bits32_reg(R3);
reg_class rarg2_bits32_reg(R4);
reg_class rarg3_bits32_reg(R5);
reg_class rarg4_bits32_reg(R6);
// ----------------------------
// 64 Bit Register Classes
// ----------------------------
// 64-bit build means 64-bit pointers means hi/lo pairs
reg_class rscratch1_bits64_reg(R11_H, R11);
reg_class rscratch2_bits64_reg(R12_H, R12);
reg_class rarg1_bits64_reg(R3_H, R3);
reg_class rarg2_bits64_reg(R4_H, R4);
reg_class rarg3_bits64_reg(R5_H, R5);
reg_class rarg4_bits64_reg(R6_H, R6);
// Thread register, 'written' by tlsLoadP, see there.
reg_class thread_bits64_reg(R16_H, R16);
reg_class r19_bits64_reg(R19_H, R19);
// 64 bit registers that can be read and written i.e. these registers
// can be dest (or src) of normal instructions.
reg_class bits64_reg_rw(
/*R0_H, R0*/ // R0
/*R1_H, R1*/ // SP
R2_H, R2, // TOC
R3_H, R3,
R4_H, R4,
R5_H, R5,
R6_H, R6,
R7_H, R7,
R8_H, R8,
R9_H, R9,
R10_H, R10,
R11_H, R11,
R12_H, R12,
/*R13_H, R13*/ // system thread id
R14_H, R14,
R15_H, R15,
/*R16_H, R16*/ // R16_thread
R17_H, R17,
R18_H, R18,
R19_H, R19,
R20_H, R20,
R21_H, R21,
R22_H, R22,
R23_H, R23,
R24_H, R24,
R25_H, R25,
R26_H, R26,
R27_H, R27,
R28_H, R28,
/*R29_H, R29,*/
R30_H, R30,
R31_H, R31
);
// 64 bit registers used excluding r2, r11 and r12
// Used to hold the TOC to avoid collisions with expanded LeafCall which uses
// r2, r11 and r12 internally.
reg_class bits64_reg_leaf_call(
/*R0_H, R0*/ // R0
/*R1_H, R1*/ // SP
/*R2_H, R2*/ // TOC
R3_H, R3,
R4_H, R4,
R5_H, R5,
R6_H, R6,
R7_H, R7,
R8_H, R8,
R9_H, R9,
R10_H, R10,
/*R11_H, R11*/
/*R12_H, R12*/
/*R13_H, R13*/ // system thread id
R14_H, R14,
R15_H, R15,
/*R16_H, R16*/ // R16_thread
R17_H, R17,
R18_H, R18,
R19_H, R19,
R20_H, R20,
R21_H, R21,
R22_H, R22,
R23_H, R23,
R24_H, R24,
R25_H, R25,
R26_H, R26,
R27_H, R27,
R28_H, R28,
/*R29_H, R29,*/
R30_H, R30,
R31_H, R31
);
// Used to hold the TOC to avoid collisions with expanded DynamicCall
// which uses r19 as inline cache internally and expanded LeafCall which uses
// r2, r11 and r12 internally.
reg_class bits64_constant_table_base(
/*R0_H, R0*/ // R0
/*R1_H, R1*/ // SP
/*R2_H, R2*/ // TOC
R3_H, R3,
R4_H, R4,
R5_H, R5,
R6_H, R6,
R7_H, R7,
R8_H, R8,
R9_H, R9,
R10_H, R10,
/*R11_H, R11*/
/*R12_H, R12*/
/*R13_H, R13*/ // system thread id
R14_H, R14,
R15_H, R15,
/*R16_H, R16*/ // R16_thread
R17_H, R17,
R18_H, R18,
/*R19_H, R19*/
R20_H, R20,
R21_H, R21,
R22_H, R22,
R23_H, R23,
R24_H, R24,
R25_H, R25,
R26_H, R26,
R27_H, R27,
R28_H, R28,
/*R29_H, R29,*/
R30_H, R30,
R31_H, R31
);
// 64 bit registers that can only be read i.e. these registers can
// only be src of all instructions.
reg_class bits64_reg_ro(
/*R0_H, R0*/ // R0
R1_H, R1,
R2_H, R2, // TOC
R3_H, R3,
R4_H, R4,
R5_H, R5,
R6_H, R6,
R7_H, R7,
R8_H, R8,
R9_H, R9,
R10_H, R10,
R11_H, R11,
R12_H, R12,
/*R13_H, R13*/ // system thread id
R14_H, R14,
R15_H, R15,
R16_H, R16, // R16_thread
R17_H, R17,
R18_H, R18,
R19_H, R19,
R20_H, R20,
R21_H, R21,
R22_H, R22,
R23_H, R23,
R24_H, R24,
R25_H, R25,
R26_H, R26,
R27_H, R27,
R28_H, R28,
/*R29_H, R29,*/ // TODO: let allocator handle TOC!!
R30_H, R30,
R31_H, R31
);
// ----------------------------
// Special Class for Condition Code Flags Register
reg_class int_flags(
/*CCR0*/ // scratch
/*CCR1*/ // scratch
/*CCR2*/ // nv!
/*CCR3*/ // nv!
/*CCR4*/ // nv!
CCR5,
CCR6,
CCR7
);
reg_class int_flags_ro(
CCR0,
CCR1,
CCR2,
CCR3,
CCR4,
CCR5,
CCR6,
CCR7
);
reg_class int_flags_CR0(CCR0);
reg_class int_flags_CR1(CCR1);
reg_class int_flags_CR6(CCR6);
reg_class ctr_reg(SR_CTR);
// ----------------------------
// Float Register Classes
// ----------------------------
reg_class flt_reg(
F0,
F1,
F2,
F3,
F4,
F5,
F6,
F7,
F8,
F9,
F10,
F11,
F12,
F13,
F14, // nv!
F15, // nv!
F16, // nv!
F17, // nv!
F18, // nv!
F19, // nv!
F20, // nv!
F21, // nv!
F22, // nv!
F23, // nv!
F24, // nv!
F25, // nv!
F26, // nv!
F27, // nv!
F28, // nv!
F29, // nv!
F30, // nv!
F31 // nv!
);
// Double precision float registers have virtual `high halves' that
// are needed by the allocator.
reg_class dbl_reg(
F0, F0_H,
F1, F1_H,
F2, F2_H,
F3, F3_H,
F4, F4_H,
F5, F5_H,
F6, F6_H,
F7, F7_H,
F8, F8_H,
F9, F9_H,
F10, F10_H,
F11, F11_H,
F12, F12_H,
F13, F13_H,
F14, F14_H, // nv!
F15, F15_H, // nv!
F16, F16_H, // nv!
F17, F17_H, // nv!
F18, F18_H, // nv!
F19, F19_H, // nv!
F20, F20_H, // nv!
F21, F21_H, // nv!
F22, F22_H, // nv!
F23, F23_H, // nv!
F24, F24_H, // nv!
F25, F25_H, // nv!
F26, F26_H, // nv!
F27, F27_H, // nv!
F28, F28_H, // nv!
F29, F29_H, // nv!
F30, F30_H, // nv!
F31, F31_H // nv!
);
%}
//----------DEFINITION BLOCK---------------------------------------------------
// Define name --> value mappings to inform the ADLC of an integer valued name
// Current support includes integer values in the range [0, 0x7FFFFFFF]
// Format:
// int_def <name> ( <int_value>, <expression>);
// Generated Code in ad_<arch>.hpp
// #define <name> (<expression>)
// // value == <int_value>
// Generated code in ad_<arch>.cpp adlc_verification()
// assert( <name> == <int_value>, "Expect (<expression>) to equal <int_value>");
//
definitions %{
// The default cost (of an ALU instruction).
int_def DEFAULT_COST_LOW ( 30, 30);
int_def DEFAULT_COST ( 100, 100);
int_def HUGE_COST (1000000, 1000000);
// Memory refs
int_def MEMORY_REF_COST_LOW ( 200, DEFAULT_COST * 2);
int_def MEMORY_REF_COST ( 300, DEFAULT_COST * 3);
// Branches are even more expensive.
int_def BRANCH_COST ( 900, DEFAULT_COST * 9);
int_def CALL_COST ( 1300, DEFAULT_COST * 13);
%}
//----------SOURCE BLOCK-------------------------------------------------------
// This is a block of C++ code which provides values, functions, and
// definitions necessary in the rest of the architecture description.
source_hpp %{
// Header information of the source block.
// Method declarations/definitions which are used outside
// the ad-scope can conveniently be defined here.
//
// To keep related declarations/definitions/uses close together,
// we switch between source %{ }% and source_hpp %{ }% freely as needed.
// Returns true if Node n is followed by a MemBar node that
// will do an acquire. If so, this node must not do the acquire
// operation.
bool followed_by_acquire(const Node *n);
%}
source %{
// Optimize load-acquire.
//
// Check if acquire is unnecessary due to following operation that does
// acquire anyways.
// Walk the pattern:
//
// n: Load.acq
// |
// MemBarAcquire
// | |
// Proj(ctrl) Proj(mem)
// | |
// MemBarRelease/Volatile
//
bool followed_by_acquire(const Node *load) {
assert(load->is_Load(), "So far implemented only for loads.");
// Find MemBarAcquire.
const Node *mba = NULL;
for (DUIterator_Fast imax, i = load->fast_outs(imax); i < imax; i++) {
const Node *out = load->fast_out(i);
if (out->Opcode() == Op_MemBarAcquire) {
if (out->in(0) == load) continue; // Skip control edge, membar should be found via precedence edge.
mba = out;
break;
}
}
if (!mba) return false;
// Find following MemBar node.
//
// The following node must be reachable by control AND memory
// edge to assure no other operations are in between the two nodes.
//
// So first get the Proj node, mem_proj, to use it to iterate forward.
Node *mem_proj = NULL;
for (DUIterator_Fast imax, i = mba->fast_outs(imax); i < imax; i++) {
mem_proj = mba->fast_out(i); // Throw out-of-bounds if proj not found
assert(mem_proj->is_Proj(), "only projections here");
ProjNode *proj = mem_proj->as_Proj();
if (proj->_con == TypeFunc::Memory &&
!Compile::current()->node_arena()->contains(mem_proj)) // Unmatched old-space only
break;
}
assert(mem_proj->as_Proj()->_con == TypeFunc::Memory, "Graph broken");
// Search MemBar behind Proj. If there are other memory operations
// behind the Proj we lost.
for (DUIterator_Fast jmax, j = mem_proj->fast_outs(jmax); j < jmax; j++) {
Node *x = mem_proj->fast_out(j);
// Proj might have an edge to a store or load node which precedes the membar.
if (x->is_Mem()) return false;
// On PPC64 release and volatile are implemented by an instruction
// that also has acquire semantics. I.e. there is no need for an
// acquire before these.
int xop = x->Opcode();
if (xop == Op_MemBarRelease || xop == Op_MemBarVolatile) {
// Make sure we're not missing Call/Phi/MergeMem by checking
// control edges. The control edge must directly lead back
// to the MemBarAcquire
Node *ctrl_proj = x->in(0);
if (ctrl_proj->is_Proj() && ctrl_proj->in(0) == mba) {
return true;
}
}
}
return false;
}
#define __ _masm.
// Tertiary op of a LoadP or StoreP encoding.
#define REGP_OP true
// ****************************************************************************
// REQUIRED FUNCTIONALITY
// !!!!! Special hack to get all type of calls to specify the byte offset
// from the start of the call to the point where the return address
// will point.
// PPC port: Removed use of lazy constant construct.
int MachCallStaticJavaNode::ret_addr_offset() {
// It's only a single branch-and-link instruction.
return 4;
}
int MachCallDynamicJavaNode::ret_addr_offset() {
// Offset is 4 with postalloc expanded calls (bl is one instruction). We use
// postalloc expanded calls if we use inline caches and do not update method data.
if (UseInlineCaches)
return 4;
int vtable_index = this->_vtable_index;
if (vtable_index < 0) {
// Must be invalid_vtable_index, not nonvirtual_vtable_index.
assert(vtable_index == Method::invalid_vtable_index, "correct sentinel value");
return 12;
} else {
assert(!UseInlineCaches, "expect vtable calls only if not using ICs");
return 24;
}
}
int MachCallRuntimeNode::ret_addr_offset() {
#if defined(ABI_ELFv2)
return 28;
#else
return 40;
#endif
}
//=============================================================================
// condition code conversions
static int cc_to_boint(int cc) {
return Assembler::bcondCRbiIs0 | (cc & 8);
}
static int cc_to_inverse_boint(int cc) {
return Assembler::bcondCRbiIs0 | (8-(cc & 8));
}
static int cc_to_biint(int cc, int flags_reg) {
return (flags_reg << 2) | (cc & 3);
}
//=============================================================================
// Compute padding required for nodes which need alignment. The padding
// is the number of bytes (not instructions) which will be inserted before
// the instruction. The padding must match the size of a NOP instruction.
int string_indexOf_imm1_charNode::compute_padding(int current_offset) const {
return (3*4-current_offset)&31; // see MacroAssembler::string_indexof_1
}
int string_indexOf_imm1Node::compute_padding(int current_offset) const {
return (3*4-current_offset)&31; // see MacroAssembler::string_indexof_1
}
int string_indexOfCharNode::compute_padding(int current_offset) const {
return (3*4-current_offset)&31; // see MacroAssembler::string_indexof_1
}
int string_indexOf_immNode::compute_padding(int current_offset) const {
return (3*4-current_offset)&31; // see MacroAssembler::string_indexof(constant needlecount)
}
int string_indexOfNode::compute_padding(int current_offset) const {
return (1*4-current_offset)&31; // see MacroAssembler::string_indexof(variable needlecount)
}
int string_compareNode::compute_padding(int current_offset) const {
return (2*4-current_offset)&31; // see MacroAssembler::string_compare
}
int string_equals_immNode::compute_padding(int current_offset) const {
if (opnd_array(3)->constant() < 16) return 0; // For strlen < 16 no nops because loop completely unrolled
return (2*4-current_offset)&31; // Genral case - see MacroAssembler::char_arrays_equalsImm
}
int string_equalsNode::compute_padding(int current_offset) const {
return (7*4-current_offset)&31; // see MacroAssembler::char_arrays_equals
}
int inlineCallClearArrayNode::compute_padding(int current_offset) const {
return (2*4-current_offset)&31; // see MacroAssembler::clear_memory_doubleword
}
//=============================================================================
// Indicate if the safepoint node needs the polling page as an input.
bool SafePointNode::needs_polling_address_input() {
// The address is loaded from thread by a seperate node.
return true;
}
//=============================================================================
// Emit an interrupt that is caught by the debugger (for debugging compiler).
void emit_break(CodeBuffer &cbuf) {
MacroAssembler _masm(&cbuf);
__ illtrap();
}
#ifndef PRODUCT
void MachBreakpointNode::format(PhaseRegAlloc *ra_, outputStream *st) const {
st->print("BREAKPOINT");
}
#endif
void MachBreakpointNode::emit(CodeBuffer &cbuf, PhaseRegAlloc *ra_) const {
emit_break(cbuf);
}
uint MachBreakpointNode::size(PhaseRegAlloc *ra_) const {
return MachNode::size(ra_);
}
//=============================================================================
void emit_nop(CodeBuffer &cbuf) {
MacroAssembler _masm(&cbuf);
__ nop();
}
static inline void emit_long(CodeBuffer &cbuf, int value) {
*((int*)(cbuf.insts_end())) = value;
cbuf.set_insts_end(cbuf.insts_end() + BytesPerInstWord);
}
//=============================================================================
%} // interrupt source
source_hpp %{ // Header information of the source block.
//--------------------------------------------------------------
//---< Used for optimization in Compile::Shorten_branches >---
//--------------------------------------------------------------
class CallStubImpl {
public:
// Emit call stub, compiled java to interpreter.
static void emit_trampoline_stub(MacroAssembler &_masm, int destination_toc_offset, int insts_call_instruction_offset);
// Size of call trampoline stub.
// This doesn't need to be accurate to the byte, but it
// must be larger than or equal to the real size of the stub.
static uint size_call_trampoline() {
return MacroAssembler::trampoline_stub_size;
}
// number of relocations needed by a call trampoline stub
static uint reloc_call_trampoline() {
return 5;
}
};
%} // end source_hpp
source %{
// Emit a trampoline stub for a call to a target which is too far away.
//
// code sequences:
//
// call-site:
// branch-and-link to <destination> or <trampoline stub>
//
// Related trampoline stub for this call-site in the stub section:
// load the call target from the constant pool
// branch via CTR (LR/link still points to the call-site above)
void CallStubImpl::emit_trampoline_stub(MacroAssembler &_masm, int destination_toc_offset, int insts_call_instruction_offset) {
address stub = __ emit_trampoline_stub(destination_toc_offset, insts_call_instruction_offset);
if (stub == NULL) {
ciEnv::current()->record_out_of_memory_failure();
}
}
//=============================================================================
// Emit an inline branch-and-link call and a related trampoline stub.
//
// code sequences:
//
// call-site:
// branch-and-link to <destination> or <trampoline stub>
//
// Related trampoline stub for this call-site in the stub section:
// load the call target from the constant pool
// branch via CTR (LR/link still points to the call-site above)
//
typedef struct {
int insts_call_instruction_offset;
int ret_addr_offset;
} EmitCallOffsets;
// Emit a branch-and-link instruction that branches to a trampoline.
// - Remember the offset of the branch-and-link instruction.
// - Add a relocation at the branch-and-link instruction.
// - Emit a branch-and-link.
// - Remember the return pc offset.
EmitCallOffsets emit_call_with_trampoline_stub(MacroAssembler &_masm, address entry_point, relocInfo::relocType rtype) {
EmitCallOffsets offsets = { -1, -1 };
const int start_offset = __ offset();
offsets.insts_call_instruction_offset = __ offset();
// No entry point given, use the current pc.
if (entry_point == NULL) entry_point = __ pc();
if (!Compile::current()->in_scratch_emit_size()) {
// Put the entry point as a constant into the constant pool.
const address entry_point_toc_addr = __ address_constant(entry_point, RelocationHolder::none);
if (entry_point_toc_addr == NULL) {
ciEnv::current()->record_out_of_memory_failure();
return offsets;
}
const int entry_point_toc_offset = __ offset_to_method_toc(entry_point_toc_addr);
// Emit the trampoline stub which will be related to the branch-and-link below.
CallStubImpl::emit_trampoline_stub(_masm, entry_point_toc_offset, offsets.insts_call_instruction_offset);
if (ciEnv::current()->failing()) { return offsets; } // Code cache may be full.
__ relocate(rtype);
}
// Note: At this point we do not have the address of the trampoline
// stub, and the entry point might be too far away for bl, so __ pc()
// serves as dummy and the bl will be patched later.
__ bl((address) __ pc());
offsets.ret_addr_offset = __ offset() - start_offset;
return offsets;
}
//=============================================================================
// Factory for creating loadConL* nodes for large/small constant pool.
static inline jlong replicate_immF(float con) {
// Replicate float con 2 times and pack into vector.
int val = *((int*)&con);
jlong lval = val;
lval = (lval << 32) | (lval & 0xFFFFFFFFl);
return lval;
}
//=============================================================================
const RegMask& MachConstantBaseNode::_out_RegMask = BITS64_CONSTANT_TABLE_BASE_mask();
int Compile::ConstantTable::calculate_table_base_offset() const {
return 0; // absolute addressing, no offset
}
bool MachConstantBaseNode::requires_postalloc_expand() const { return true; }
void MachConstantBaseNode::postalloc_expand(GrowableArray <Node *> *nodes, PhaseRegAlloc *ra_) {
iRegPdstOper *op_dst = new iRegPdstOper();
MachNode *m1 = new loadToc_hiNode();
MachNode *m2 = new loadToc_loNode();
m1->add_req(NULL);
m2->add_req(NULL, m1);
m1->_opnds[0] = op_dst;
m2->_opnds[0] = op_dst;
m2->_opnds[1] = op_dst;
ra_->set_pair(m1->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
ra_->set_pair(m2->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
nodes->push(m1);
nodes->push(m2);
}
void MachConstantBaseNode::emit(CodeBuffer& cbuf, PhaseRegAlloc* ra_) const {
// Is postalloc expanded.
ShouldNotReachHere();
}
uint MachConstantBaseNode::size(PhaseRegAlloc* ra_) const {
return 0;
}
#ifndef PRODUCT
void MachConstantBaseNode::format(PhaseRegAlloc* ra_, outputStream* st) const {
st->print("-- \t// MachConstantBaseNode (empty encoding)");
}
#endif
//=============================================================================
#ifndef PRODUCT
void MachPrologNode::format(PhaseRegAlloc *ra_, outputStream *st) const {
Compile* C = ra_->C;
const long framesize = C->frame_slots() << LogBytesPerInt;
st->print("PROLOG\n\t");
if (C->need_stack_bang(framesize)) {
st->print("stack_overflow_check\n\t");
}
if (!false /* TODO: PPC port C->is_frameless_method()*/) {
st->print("save return pc\n\t");
st->print("push frame %ld\n\t", -framesize);
}
}
#endif
// Macro used instead of the common __ to emulate the pipes of PPC.
// Instead of e.g. __ ld(...) one hase to write ___(ld) ld(...) This enables the
// micro scheduler to cope with "hand written" assembler like in the prolog. Though
// still no scheduling of this code is possible, the micro scheduler is aware of the
// code and can update its internal data. The following mechanism is used to achieve this:
// The micro scheduler calls size() of each compound node during scheduling. size() does a
// dummy emit and only during this dummy emit C->hb_scheduling() is not NULL.
#if 0 // TODO: PPC port
#define ___(op) if (UsePower6SchedulerPPC64 && C->hb_scheduling()) \
C->hb_scheduling()->_pdScheduling->PdEmulatePipe(ppc64Opcode_##op); \
_masm.
#define ___stop if (UsePower6SchedulerPPC64 && C->hb_scheduling()) \
C->hb_scheduling()->_pdScheduling->PdEmulatePipe(archOpcode_none)
#define ___advance if (UsePower6SchedulerPPC64 && C->hb_scheduling()) \
C->hb_scheduling()->_pdScheduling->advance_offset
#else
#define ___(op) if (UsePower6SchedulerPPC64) \
Unimplemented(); \
_masm.
#define ___stop if (UsePower6SchedulerPPC64) \
Unimplemented()
#define ___advance if (UsePower6SchedulerPPC64) \
Unimplemented()
#endif
void MachPrologNode::emit(CodeBuffer &cbuf, PhaseRegAlloc *ra_) const {
Compile* C = ra_->C;
MacroAssembler _masm(&cbuf);
const long framesize = C->frame_size_in_bytes();
assert(framesize % (2 * wordSize) == 0, "must preserve 2*wordSize alignment");
const bool method_is_frameless = false /* TODO: PPC port C->is_frameless_method()*/;
const Register return_pc = R20; // Must match return_addr() in frame section.
const Register callers_sp = R21;
const Register push_frame_temp = R22;
const Register toc_temp = R23;
assert_different_registers(R11, return_pc, callers_sp, push_frame_temp, toc_temp);
if (method_is_frameless) {
// Add nop at beginning of all frameless methods to prevent any
// oop instructions from getting overwritten by make_not_entrant
// (patching attempt would fail).
___(nop) nop();
} else {
// Get return pc.
___(mflr) mflr(return_pc);
}
// Calls to C2R adapters often do not accept exceptional returns.
// We require that their callers must bang for them. But be
// careful, because some VM calls (such as call site linkage) can
// use several kilobytes of stack. But the stack safety zone should
// account for that. See bugs 4446381, 4468289, 4497237.
int bangsize = C->bang_size_in_bytes();
assert(bangsize >= framesize || bangsize <= 0, "stack bang size incorrect");
if (C->need_stack_bang(bangsize) && UseStackBanging) {
// Unfortunately we cannot use the function provided in
// assembler.cpp as we have to emulate the pipes. So I had to
// insert the code of generate_stack_overflow_check(), see
// assembler.cpp for some illuminative comments.
const int page_size = os::vm_page_size();
int bang_end = JavaThread::stack_shadow_zone_size();
// This is how far the previous frame's stack banging extended.
const int bang_end_safe = bang_end;
if (bangsize > page_size) {
bang_end += bangsize;
}
int bang_offset = bang_end_safe;
while (bang_offset <= bang_end) {
// Need at least one stack bang at end of shadow zone.
// Again I had to copy code, this time from assembler_ppc.cpp,
// bang_stack_with_offset - see there for comments.
// Stack grows down, caller passes positive offset.
assert(bang_offset > 0, "must bang with positive offset");
long stdoffset = -bang_offset;
if (Assembler::is_simm(stdoffset, 16)) {
// Signed 16 bit offset, a simple std is ok.
if (UseLoadInstructionsForStackBangingPPC64) {
___(ld) ld(R0, (int)(signed short)stdoffset, R1_SP);
} else {
___(std) std(R0, (int)(signed short)stdoffset, R1_SP);
}
} else if (Assembler::is_simm(stdoffset, 31)) {
// Use largeoffset calculations for addis & ld/std.
const int hi = MacroAssembler::largeoffset_si16_si16_hi(stdoffset);
const int lo = MacroAssembler::largeoffset_si16_si16_lo(stdoffset);
Register tmp = R11;
___(addis) addis(tmp, R1_SP, hi);
if (UseLoadInstructionsForStackBangingPPC64) {
___(ld) ld(R0, lo, tmp);
} else {
___(std) std(R0, lo, tmp);
}
} else {
ShouldNotReachHere();
}
bang_offset += page_size;
}
// R11 trashed
} // C->need_stack_bang(framesize) && UseStackBanging
unsigned int bytes = (unsigned int)framesize;
long offset = Assembler::align_addr(bytes, frame::alignment_in_bytes);
ciMethod *currMethod = C->method();
// Optimized version for most common case.
if (UsePower6SchedulerPPC64 &&
!method_is_frameless && Assembler::is_simm((int)(-offset), 16) &&
!(false /* ConstantsALot TODO: PPC port*/)) {
___(or) mr(callers_sp, R1_SP);
___(std) std(return_pc, _abi(lr), R1_SP);
___(stdu) stdu(R1_SP, -offset, R1_SP);
return;
}
if (!method_is_frameless) {
// Get callers sp.
___(or) mr(callers_sp, R1_SP);
// Push method's frame, modifies SP.
assert(Assembler::is_uimm(framesize, 32U), "wrong type");
// The ABI is already accounted for in 'framesize' via the
// 'out_preserve' area.
Register tmp = push_frame_temp;
// Had to insert code of push_frame((unsigned int)framesize, push_frame_temp).
if (Assembler::is_simm(-offset, 16)) {
___(stdu) stdu(R1_SP, -offset, R1_SP);
} else {
long x = -offset;
// Had to insert load_const(tmp, -offset).
___(addis) lis( tmp, (int)((signed short)(((x >> 32) & 0xffff0000) >> 16)));
___(ori) ori( tmp, tmp, ((x >> 32) & 0x0000ffff));
___(rldicr) sldi(tmp, tmp, 32);
___(oris) oris(tmp, tmp, (x & 0xffff0000) >> 16);
___(ori) ori( tmp, tmp, (x & 0x0000ffff));
___(stdux) stdux(R1_SP, R1_SP, tmp);
}
}
#if 0 // TODO: PPC port
// For testing large constant pools, emit a lot of constants to constant pool.
// "Randomize" const_size.
if (ConstantsALot) {
const int num_consts = const_size();
for (int i = 0; i < num_consts; i++) {
__ long_constant(0xB0B5B00BBABE);
}
}
#endif
if (!method_is_frameless) {
// Save return pc.
___(std) std(return_pc, _abi(lr), callers_sp);
}
}
#undef ___
#undef ___stop
#undef ___advance
uint MachPrologNode::size(PhaseRegAlloc *ra_) const {
// Variable size. determine dynamically.
return MachNode::size(ra_);
}
int MachPrologNode::reloc() const {
// Return number of relocatable values contained in this instruction.
return 1; // 1 reloc entry for load_const(toc).
}
//=============================================================================
#ifndef PRODUCT
void MachEpilogNode::format(PhaseRegAlloc *ra_, outputStream *st) const {
Compile* C = ra_->C;
st->print("EPILOG\n\t");
st->print("restore return pc\n\t");
st->print("pop frame\n\t");
if (do_polling() && C->is_method_compilation()) {
st->print("touch polling page\n\t");
}
}
#endif
void MachEpilogNode::emit(CodeBuffer &cbuf, PhaseRegAlloc *ra_) const {
Compile* C = ra_->C;
MacroAssembler _masm(&cbuf);
const long framesize = ((long)C->frame_slots()) << LogBytesPerInt;
assert(framesize >= 0, "negative frame-size?");
const bool method_needs_polling = do_polling() && C->is_method_compilation();
const bool method_is_frameless = false /* TODO: PPC port C->is_frameless_method()*/;
const Register return_pc = R11;
const Register polling_page = R12;
if (!method_is_frameless) {
// Restore return pc relative to callers' sp.
__ ld(return_pc, ((int)framesize) + _abi(lr), R1_SP);
}
if (method_needs_polling) {
if (LoadPollAddressFromThread) {
// TODO: PPC port __ ld(polling_page, in_bytes(JavaThread::poll_address_offset()), R16_thread);
Unimplemented();
} else {
__ load_const_optimized(polling_page, (long)(address) os::get_polling_page()); // TODO: PPC port: get_standard_polling_page()
}
}
if (!method_is_frameless) {
// Move return pc to LR.
__ mtlr(return_pc);
// Pop frame (fixed frame-size).
__ addi(R1_SP, R1_SP, (int)framesize);
}
if (method_needs_polling) {
// We need to mark the code position where the load from the safepoint
// polling page was emitted as relocInfo::poll_return_type here.
__ relocate(relocInfo::poll_return_type);
__ load_from_polling_page(polling_page);
}
}
uint MachEpilogNode::size(PhaseRegAlloc *ra_) const {
// Variable size. Determine dynamically.
return MachNode::size(ra_);
}
int MachEpilogNode::reloc() const {
// Return number of relocatable values contained in this instruction.
return 1; // 1 for load_from_polling_page.
}
const Pipeline * MachEpilogNode::pipeline() const {
return MachNode::pipeline_class();
}
// This method seems to be obsolete. It is declared in machnode.hpp
// and defined in all *.ad files, but it is never called. Should we
// get rid of it?
int MachEpilogNode::safepoint_offset() const {
assert(do_polling(), "no return for this epilog node");
return 0;
}
#if 0 // TODO: PPC port
void MachLoadPollAddrLateNode::emit(CodeBuffer& cbuf, PhaseRegAlloc* ra_) const {
MacroAssembler _masm(&cbuf);
if (LoadPollAddressFromThread) {
_masm.ld(R11, in_bytes(JavaThread::poll_address_offset()), R16_thread);
} else {
_masm.nop();
}
}
uint MachLoadPollAddrLateNode::size(PhaseRegAlloc* ra_) const {
if (LoadPollAddressFromThread) {
return 4;
} else {
return 4;
}
}
#ifndef PRODUCT
void MachLoadPollAddrLateNode::format(PhaseRegAlloc* ra_, outputStream* st) const {
st->print_cr(" LD R11, PollAddressOffset, R16_thread \t// LoadPollAddressFromThread");
}
#endif
const RegMask &MachLoadPollAddrLateNode::out_RegMask() const {
return RSCRATCH1_BITS64_REG_mask();
}
#endif // PPC port
// =============================================================================
// Figure out which register class each belongs in: rc_int, rc_float or
// rc_stack.
enum RC { rc_bad, rc_int, rc_float, rc_stack };
static enum RC rc_class(OptoReg::Name reg) {
// Return the register class for the given register. The given register
// reg is a <register>_num value, which is an index into the MachRegisterNumbers
// enumeration in adGlobals_ppc.hpp.
if (reg == OptoReg::Bad) return rc_bad;
// We have 64 integer register halves, starting at index 0.
if (reg < 64) return rc_int;
// We have 64 floating-point register halves, starting at index 64.
if (reg < 64+64) return rc_float;
// Between float regs & stack are the flags regs.
assert(OptoReg::is_stack(reg), "blow up if spilling flags");
return rc_stack;
}
static int ld_st_helper(CodeBuffer *cbuf, const char *op_str, uint opcode, int reg, int offset,
bool do_print, Compile* C, outputStream *st) {
assert(opcode == Assembler::LD_OPCODE ||
opcode == Assembler::STD_OPCODE ||
opcode == Assembler::LWZ_OPCODE ||
opcode == Assembler::STW_OPCODE ||
opcode == Assembler::LFD_OPCODE ||
opcode == Assembler::STFD_OPCODE ||
opcode == Assembler::LFS_OPCODE ||
opcode == Assembler::STFS_OPCODE,
"opcode not supported");
if (cbuf) {
int d =
(Assembler::LD_OPCODE == opcode || Assembler::STD_OPCODE == opcode) ?
Assembler::ds(offset+0 /* TODO: PPC port C->frame_slots_sp_bias_in_bytes()*/)
: Assembler::d1(offset+0 /* TODO: PPC port C->frame_slots_sp_bias_in_bytes()*/); // Makes no difference in opt build.
emit_long(*cbuf, opcode | Assembler::rt(Matcher::_regEncode[reg]) | d | Assembler::ra(R1_SP));
}
#ifndef PRODUCT
else if (do_print) {
st->print("%-7s %s, [R1_SP + #%d+%d] \t// spill copy",
op_str,
Matcher::regName[reg],
offset, 0 /* TODO: PPC port C->frame_slots_sp_bias_in_bytes()*/);
}
#endif
return 4; // size
}
uint MachSpillCopyNode::implementation(CodeBuffer *cbuf, PhaseRegAlloc *ra_, bool do_size, outputStream *st) const {
Compile* C = ra_->C;
// Get registers to move.
OptoReg::Name src_hi = ra_->get_reg_second(in(1));
OptoReg::Name src_lo = ra_->get_reg_first(in(1));
OptoReg::Name dst_hi = ra_->get_reg_second(this);
OptoReg::Name dst_lo = ra_->get_reg_first(this);
enum RC src_hi_rc = rc_class(src_hi);
enum RC src_lo_rc = rc_class(src_lo);
enum RC dst_hi_rc = rc_class(dst_hi);
enum RC dst_lo_rc = rc_class(dst_lo);
assert(src_lo != OptoReg::Bad && dst_lo != OptoReg::Bad, "must move at least 1 register");
if (src_hi != OptoReg::Bad)
assert((src_lo&1)==0 && src_lo+1==src_hi &&
(dst_lo&1)==0 && dst_lo+1==dst_hi,
"expected aligned-adjacent pairs");
// Generate spill code!
int size = 0;
if (src_lo == dst_lo && src_hi == dst_hi)
return size; // Self copy, no move.
// --------------------------------------
// Memory->Memory Spill. Use R0 to hold the value.
if (src_lo_rc == rc_stack && dst_lo_rc == rc_stack) {
int src_offset = ra_->reg2offset(src_lo);
int dst_offset = ra_->reg2offset(dst_lo);
if (src_hi != OptoReg::Bad) {
assert(src_hi_rc==rc_stack && dst_hi_rc==rc_stack,
"expected same type of move for high parts");
size += ld_st_helper(cbuf, "LD ", Assembler::LD_OPCODE, R0_num, src_offset, !do_size, C, st);
if (!cbuf && !do_size) st->print("\n\t");
size += ld_st_helper(cbuf, "STD ", Assembler::STD_OPCODE, R0_num, dst_offset, !do_size, C, st);
} else {
size += ld_st_helper(cbuf, "LWZ ", Assembler::LWZ_OPCODE, R0_num, src_offset, !do_size, C, st);
if (!cbuf && !do_size) st->print("\n\t");
size += ld_st_helper(cbuf, "STW ", Assembler::STW_OPCODE, R0_num, dst_offset, !do_size, C, st);
}
return size;
}
// --------------------------------------
// Check for float->int copy; requires a trip through memory.
if (src_lo_rc == rc_float && dst_lo_rc == rc_int) {
Unimplemented();
}
// --------------------------------------
// Check for integer reg-reg copy.
if (src_lo_rc == rc_int && dst_lo_rc == rc_int) {
Register Rsrc = as_Register(Matcher::_regEncode[src_lo]);
Register Rdst = as_Register(Matcher::_regEncode[dst_lo]);
size = (Rsrc != Rdst) ? 4 : 0;
if (cbuf) {
MacroAssembler _masm(cbuf);
if (size) {
__ mr(Rdst, Rsrc);
}
}
#ifndef PRODUCT
else if (!do_size) {
if (size) {
st->print("%-7s %s, %s \t// spill copy", "MR", Matcher::regName[dst_lo], Matcher::regName[src_lo]);
} else {
st->print("%-7s %s, %s \t// spill copy", "MR-NOP", Matcher::regName[dst_lo], Matcher::regName[src_lo]);
}
}
#endif
return size;
}
// Check for integer store.
if (src_lo_rc == rc_int && dst_lo_rc == rc_stack) {
int dst_offset = ra_->reg2offset(dst_lo);
if (src_hi != OptoReg::Bad) {
assert(src_hi_rc==rc_int && dst_hi_rc==rc_stack,
"expected same type of move for high parts");
size += ld_st_helper(cbuf, "STD ", Assembler::STD_OPCODE, src_lo, dst_offset, !do_size, C, st);
} else {
size += ld_st_helper(cbuf, "STW ", Assembler::STW_OPCODE, src_lo, dst_offset, !do_size, C, st);
}
return size;
}
// Check for integer load.
if (dst_lo_rc == rc_int && src_lo_rc == rc_stack) {
int src_offset = ra_->reg2offset(src_lo);
if (src_hi != OptoReg::Bad) {
assert(dst_hi_rc==rc_int && src_hi_rc==rc_stack,
"expected same type of move for high parts");
size += ld_st_helper(cbuf, "LD ", Assembler::LD_OPCODE, dst_lo, src_offset, !do_size, C, st);
} else {
size += ld_st_helper(cbuf, "LWZ ", Assembler::LWZ_OPCODE, dst_lo, src_offset, !do_size, C, st);
}
return size;
}
// Check for float reg-reg copy.
if (src_lo_rc == rc_float && dst_lo_rc == rc_float) {
if (cbuf) {
MacroAssembler _masm(cbuf);
FloatRegister Rsrc = as_FloatRegister(Matcher::_regEncode[src_lo]);
FloatRegister Rdst = as_FloatRegister(Matcher::_regEncode[dst_lo]);
__ fmr(Rdst, Rsrc);
}
#ifndef PRODUCT
else if (!do_size) {
st->print("%-7s %s, %s \t// spill copy", "FMR", Matcher::regName[dst_lo], Matcher::regName[src_lo]);
}
#endif
return 4;
}
// Check for float store.
if (src_lo_rc == rc_float && dst_lo_rc == rc_stack) {
int dst_offset = ra_->reg2offset(dst_lo);
if (src_hi != OptoReg::Bad) {
assert(src_hi_rc==rc_float && dst_hi_rc==rc_stack,
"expected same type of move for high parts");
size += ld_st_helper(cbuf, "STFD", Assembler::STFD_OPCODE, src_lo, dst_offset, !do_size, C, st);
} else {
size += ld_st_helper(cbuf, "STFS", Assembler::STFS_OPCODE, src_lo, dst_offset, !do_size, C, st);
}
return size;
}
// Check for float load.
if (dst_lo_rc == rc_float && src_lo_rc == rc_stack) {
int src_offset = ra_->reg2offset(src_lo);
if (src_hi != OptoReg::Bad) {
assert(dst_hi_rc==rc_float && src_hi_rc==rc_stack,
"expected same type of move for high parts");
size += ld_st_helper(cbuf, "LFD ", Assembler::LFD_OPCODE, dst_lo, src_offset, !do_size, C, st);
} else {
size += ld_st_helper(cbuf, "LFS ", Assembler::LFS_OPCODE, dst_lo, src_offset, !do_size, C, st);
}
return size;
}
// --------------------------------------------------------------------
// Check for hi bits still needing moving. Only happens for misaligned
// arguments to native calls.
if (src_hi == dst_hi)
return size; // Self copy; no move.
assert(src_hi_rc != rc_bad && dst_hi_rc != rc_bad, "src_hi & dst_hi cannot be Bad");
ShouldNotReachHere(); // Unimplemented
return 0;
}
#ifndef PRODUCT
void MachSpillCopyNode::format(PhaseRegAlloc *ra_, outputStream *st) const {
if (!ra_)
st->print("N%d = SpillCopy(N%d)", _idx, in(1)->_idx);
else
implementation(NULL, ra_, false, st);
}
#endif
void MachSpillCopyNode::emit(CodeBuffer &cbuf, PhaseRegAlloc *ra_) const {
implementation(&cbuf, ra_, false, NULL);
}
uint MachSpillCopyNode::size(PhaseRegAlloc *ra_) const {
return implementation(NULL, ra_, true, NULL);
}
#if 0 // TODO: PPC port
ArchOpcode MachSpillCopyNode_archOpcode(MachSpillCopyNode *n, PhaseRegAlloc *ra_) {
#ifndef PRODUCT
if (ra_->node_regs_max_index() == 0) return archOpcode_undefined;
#endif
assert(ra_->node_regs_max_index() != 0, "");
// Get registers to move.
OptoReg::Name src_hi = ra_->get_reg_second(n->in(1));
OptoReg::Name src_lo = ra_->get_reg_first(n->in(1));
OptoReg::Name dst_hi = ra_->get_reg_second(n);
OptoReg::Name dst_lo = ra_->get_reg_first(n);
enum RC src_lo_rc = rc_class(src_lo);
enum RC dst_lo_rc = rc_class(dst_lo);
if (src_lo == dst_lo && src_hi == dst_hi)
return ppc64Opcode_none; // Self copy, no move.
// --------------------------------------
// Memory->Memory Spill. Use R0 to hold the value.
if (src_lo_rc == rc_stack && dst_lo_rc == rc_stack) {
return ppc64Opcode_compound;
}
// --------------------------------------
// Check for float->int copy; requires a trip through memory.
if (src_lo_rc == rc_float && dst_lo_rc == rc_int) {
Unimplemented();
}
// --------------------------------------
// Check for integer reg-reg copy.
if (src_lo_rc == rc_int && dst_lo_rc == rc_int) {
Register Rsrc = as_Register(Matcher::_regEncode[src_lo]);
Register Rdst = as_Register(Matcher::_regEncode[dst_lo]);
if (Rsrc == Rdst) {
return ppc64Opcode_none;
} else {
return ppc64Opcode_or;
}
}
// Check for integer store.
if (src_lo_rc == rc_int && dst_lo_rc == rc_stack) {
if (src_hi != OptoReg::Bad) {
return ppc64Opcode_std;
} else {
return ppc64Opcode_stw;
}
}
// Check for integer load.
if (dst_lo_rc == rc_int && src_lo_rc == rc_stack) {
if (src_hi != OptoReg::Bad) {
return ppc64Opcode_ld;
} else {
return ppc64Opcode_lwz;
}
}
// Check for float reg-reg copy.
if (src_lo_rc == rc_float && dst_lo_rc == rc_float) {
return ppc64Opcode_fmr;
}
// Check for float store.
if (src_lo_rc == rc_float && dst_lo_rc == rc_stack) {
if (src_hi != OptoReg::Bad) {
return ppc64Opcode_stfd;
} else {
return ppc64Opcode_stfs;
}
}
// Check for float load.
if (dst_lo_rc == rc_float && src_lo_rc == rc_stack) {
if (src_hi != OptoReg::Bad) {
return ppc64Opcode_lfd;
} else {
return ppc64Opcode_lfs;
}
}
// --------------------------------------------------------------------
// Check for hi bits still needing moving. Only happens for misaligned
// arguments to native calls.
if (src_hi == dst_hi) {
return ppc64Opcode_none; // Self copy; no move.
}
ShouldNotReachHere();
return ppc64Opcode_undefined;
}
#endif // PPC port
#ifndef PRODUCT
void MachNopNode::format(PhaseRegAlloc *ra_, outputStream *st) const {
st->print("NOP \t// %d nops to pad for loops.", _count);
}
#endif
void MachNopNode::emit(CodeBuffer &cbuf, PhaseRegAlloc *) const {
MacroAssembler _masm(&cbuf);
// _count contains the number of nops needed for padding.
for (int i = 0; i < _count; i++) {
__ nop();
}
}
uint MachNopNode::size(PhaseRegAlloc *ra_) const {
return _count * 4;
}
#ifndef PRODUCT
void BoxLockNode::format(PhaseRegAlloc *ra_, outputStream *st) const {
int offset = ra_->reg2offset(in_RegMask(0).find_first_elem());
char reg_str[128];
ra_->dump_register(this, reg_str);
st->print("ADDI %s, SP, %d \t// box node", reg_str, offset);
}
#endif
void BoxLockNode::emit(CodeBuffer &cbuf, PhaseRegAlloc *ra_) const {
MacroAssembler _masm(&cbuf);
int offset = ra_->reg2offset(in_RegMask(0).find_first_elem());
int reg = ra_->get_encode(this);
if (Assembler::is_simm(offset, 16)) {
__ addi(as_Register(reg), R1, offset);
} else {
ShouldNotReachHere();
}
}
uint BoxLockNode::size(PhaseRegAlloc *ra_) const {
// BoxLockNode is not a MachNode, so we can't just call MachNode::size(ra_).
return 4;
}
#ifndef PRODUCT
void MachUEPNode::format(PhaseRegAlloc *ra_, outputStream *st) const {
st->print_cr("---- MachUEPNode ----");
st->print_cr("...");
}
#endif
void MachUEPNode::emit(CodeBuffer &cbuf, PhaseRegAlloc *ra_) const {
// This is the unverified entry point.
MacroAssembler _masm(&cbuf);
// Inline_cache contains a klass.
Register ic_klass = as_Register(Matcher::inline_cache_reg_encode());
Register receiver_klass = R12_scratch2; // tmp
assert_different_registers(ic_klass, receiver_klass, R11_scratch1, R3_ARG1);
assert(R11_scratch1 == R11, "need prologue scratch register");
// Check for NULL argument if we don't have implicit null checks.
if (!ImplicitNullChecks || !os::zero_page_read_protected()) {
if (TrapBasedNullChecks) {
__ trap_null_check(R3_ARG1);
} else {
Label valid;
__ cmpdi(CCR0, R3_ARG1, 0);
__ bne_predict_taken(CCR0, valid);
// We have a null argument, branch to ic_miss_stub.
__ b64_patchable((address)SharedRuntime::get_ic_miss_stub(),
relocInfo::runtime_call_type);
__ bind(valid);
}
}
// Assume argument is not NULL, load klass from receiver.
__ load_klass(receiver_klass, R3_ARG1);
if (TrapBasedICMissChecks) {
__ trap_ic_miss_check(receiver_klass, ic_klass);
} else {
Label valid;
__ cmpd(CCR0, receiver_klass, ic_klass);
__ beq_predict_taken(CCR0, valid);
// We have an unexpected klass, branch to ic_miss_stub.
__ b64_patchable((address)SharedRuntime::get_ic_miss_stub(),
relocInfo::runtime_call_type);
__ bind(valid);
}
// Argument is valid and klass is as expected, continue.
}
#if 0 // TODO: PPC port
// Optimize UEP code on z (save a load_const() call in main path).
int MachUEPNode::ep_offset() {
return 0;
}
#endif
uint MachUEPNode::size(PhaseRegAlloc *ra_) const {
// Variable size. Determine dynamically.
return MachNode::size(ra_);
}
//=============================================================================
%} // interrupt source
source_hpp %{ // Header information of the source block.
class HandlerImpl {
public:
static int emit_exception_handler(CodeBuffer &cbuf);
static int emit_deopt_handler(CodeBuffer& cbuf);
static uint size_exception_handler() {
// The exception_handler is a b64_patchable.
return MacroAssembler::b64_patchable_size;
}
static uint size_deopt_handler() {
// The deopt_handler is a bl64_patchable.
return MacroAssembler::bl64_patchable_size;
}
};
%} // end source_hpp
source %{
int HandlerImpl::emit_exception_handler(CodeBuffer &cbuf) {
MacroAssembler _masm(&cbuf);
address base = __ start_a_stub(size_exception_handler());
if (base == NULL) return 0; // CodeBuffer::expand failed
int offset = __ offset();
__ b64_patchable((address)OptoRuntime::exception_blob()->content_begin(),
relocInfo::runtime_call_type);
assert(__ offset() - offset == (int)size_exception_handler(), "must be fixed size");
__ end_a_stub();
return offset;
}
// The deopt_handler is like the exception handler, but it calls to
// the deoptimization blob instead of jumping to the exception blob.
int HandlerImpl::emit_deopt_handler(CodeBuffer& cbuf) {
MacroAssembler _masm(&cbuf);
address base = __ start_a_stub(size_deopt_handler());
if (base == NULL) return 0; // CodeBuffer::expand failed
int offset = __ offset();
__ bl64_patchable((address)SharedRuntime::deopt_blob()->unpack(),
relocInfo::runtime_call_type);
assert(__ offset() - offset == (int) size_deopt_handler(), "must be fixed size");
__ end_a_stub();
return offset;
}
//=============================================================================
// Use a frame slots bias for frameless methods if accessing the stack.
static int frame_slots_bias(int reg_enc, PhaseRegAlloc* ra_) {
if (as_Register(reg_enc) == R1_SP) {
return 0; // TODO: PPC port ra_->C->frame_slots_sp_bias_in_bytes();
}
return 0;
}
const bool Matcher::match_rule_supported(int opcode) {
if (!has_match_rule(opcode))
return false;
switch (opcode) {
case Op_SqrtD:
return VM_Version::has_fsqrt();
case Op_CountLeadingZerosI:
case Op_CountLeadingZerosL:
case Op_CountTrailingZerosI:
case Op_CountTrailingZerosL:
if (!UseCountLeadingZerosInstructionsPPC64)
return false;
break;
case Op_PopCountI:
case Op_PopCountL:
return (UsePopCountInstruction && VM_Version::has_popcntw());
case Op_StrComp:
return SpecialStringCompareTo && !CompactStrings;
case Op_StrEquals:
return SpecialStringEquals && !CompactStrings;
case Op_StrIndexOf:
return SpecialStringIndexOf && !CompactStrings;
case Op_StrIndexOfChar:
return SpecialStringIndexOf && !CompactStrings;
}
return true; // Per default match rules are supported.
}
const bool Matcher::match_rule_supported_vector(int opcode, int vlen) {
// TODO
// identify extra cases that we might want to provide match rules for
// e.g. Op_ vector nodes and other intrinsics while guarding with vlen
bool ret_value = match_rule_supported(opcode);
// Add rules here.
return ret_value; // Per default match rules are supported.
}
const int Matcher::float_pressure(int default_pressure_threshold) {
return default_pressure_threshold;
}
int Matcher::regnum_to_fpu_offset(int regnum) {
// No user for this method?
Unimplemented();
return 999;
}
const bool Matcher::convL2FSupported(void) {
// fcfids can do the conversion (>= Power7).
// fcfid + frsp showed rounding problem when result should be 0x3f800001.
return VM_Version::has_fcfids(); // False means that conversion is done by runtime call.
}
// Vector width in bytes.
const int Matcher::vector_width_in_bytes(BasicType bt) {
assert(MaxVectorSize == 8, "");
return 8;
}
// Vector ideal reg.
const int Matcher::vector_ideal_reg(int size) {
assert(MaxVectorSize == 8 && size == 8, "");
return Op_RegL;
}
const int Matcher::vector_shift_count_ideal_reg(int size) {
fatal("vector shift is not supported");
return Node::NotAMachineReg;
}
// Limits on vector size (number of elements) loaded into vector.
const int Matcher::max_vector_size(const BasicType bt) {
assert(is_java_primitive(bt), "only primitive type vectors");
return vector_width_in_bytes(bt)/type2aelembytes(bt);
}
const int Matcher::min_vector_size(const BasicType bt) {
return max_vector_size(bt); // Same as max.
}
// PPC doesn't support misaligned vectors store/load.
const bool Matcher::misaligned_vectors_ok() {
return false;
}
// PPC AES support not yet implemented
const bool Matcher::pass_original_key_for_aes() {
return false;
}
// RETURNS: whether this branch offset is short enough that a short
// branch can be used.
//
// If the platform does not provide any short branch variants, then
// this method should return `false' for offset 0.
//
// `Compile::Fill_buffer' will decide on basis of this information
// whether to do the pass `Compile::Shorten_branches' at all.
//
// And `Compile::Shorten_branches' will decide on basis of this
// information whether to replace particular branch sites by short
// ones.
bool Matcher::is_short_branch_offset(int rule, int br_size, int offset) {
// Is the offset within the range of a ppc64 pc relative branch?
bool b;
const int safety_zone = 3 * BytesPerInstWord;
b = Assembler::is_simm((offset<0 ? offset-safety_zone : offset+safety_zone),
29 - 16 + 1 + 2);
return b;
}
const bool Matcher::isSimpleConstant64(jlong value) {
// Probably always true, even if a temp register is required.
return true;
}
/* TODO: PPC port
// Make a new machine dependent decode node (with its operands).
MachTypeNode *Matcher::make_decode_node() {
assert(Universe::narrow_oop_base() == NULL && Universe::narrow_oop_shift() == 0,
"This method is only implemented for unscaled cOops mode so far");
MachTypeNode *decode = new decodeN_unscaledNode();
decode->set_opnd_array(0, new iRegPdstOper());
decode->set_opnd_array(1, new iRegNsrcOper());
return decode;
}
*/
// Threshold size for cleararray.
const int Matcher::init_array_short_size = 8 * BytesPerLong;
// false => size gets scaled to BytesPerLong, ok.
const bool Matcher::init_array_count_is_in_bytes = false;
// Use conditional move (CMOVL) on Power7.
const int Matcher::long_cmove_cost() { return 0; } // this only makes long cmoves more expensive than int cmoves
// Suppress CMOVF. Conditional move available (sort of) on PPC64 only from P7 onwards. Not exploited yet.
// fsel doesn't accept a condition register as input, so this would be slightly different.
const int Matcher::float_cmove_cost() { return ConditionalMoveLimit; }
// Power6 requires postalloc expand (see block.cpp for description of postalloc expand).
const bool Matcher::require_postalloc_expand = true;
// Should the Matcher clone shifts on addressing modes, expecting them to
// be subsumed into complex addressing expressions or compute them into
// registers? True for Intel but false for most RISCs.
const bool Matcher::clone_shift_expressions = false;
// Do we need to mask the count passed to shift instructions or does
// the cpu only look at the lower 5/6 bits anyway?
// PowerPC requires masked shift counts.
const bool Matcher::need_masked_shift_count = true;
// This affects two different things:
// - how Decode nodes are matched
// - how ImplicitNullCheck opportunities are recognized
// If true, the matcher will try to remove all Decodes and match them
// (as operands) into nodes. NullChecks are not prepared to deal with
// Decodes by final_graph_reshaping().
// If false, final_graph_reshaping() forces the decode behind the Cmp
// for a NullCheck. The matcher matches the Decode node into a register.
// Implicit_null_check optimization moves the Decode along with the
// memory operation back up before the NullCheck.
bool Matcher::narrow_oop_use_complex_address() {
// TODO: PPC port if (MatchDecodeNodes) return true;
return false;
}
bool Matcher::narrow_klass_use_complex_address() {
NOT_LP64(ShouldNotCallThis());
assert(UseCompressedClassPointers, "only for compressed klass code");
// TODO: PPC port if (MatchDecodeNodes) return true;
return false;
}
// Is it better to copy float constants, or load them directly from memory?
// Intel can load a float constant from a direct address, requiring no
// extra registers. Most RISCs will have to materialize an address into a
// register first, so they would do better to copy the constant from stack.
const bool Matcher::rematerialize_float_constants = false;
// If CPU can load and store mis-aligned doubles directly then no fixup is
// needed. Else we split the double into 2 integer pieces and move it
// piece-by-piece. Only happens when passing doubles into C code as the
// Java calling convention forces doubles to be aligned.
const bool Matcher::misaligned_doubles_ok = true;
void Matcher::pd_implicit_null_fixup(MachNode *node, uint idx) {
Unimplemented();
}
// Advertise here if the CPU requires explicit rounding operations
// to implement the UseStrictFP mode.
const bool Matcher::strict_fp_requires_explicit_rounding = false;
// Do floats take an entire double register or just half?
//
// A float occupies a ppc64 double register. For the allocator, a
// ppc64 double register appears as a pair of float registers.
bool Matcher::float_in_double() { return true; }
// Do ints take an entire long register or just half?
// The relevant question is how the int is callee-saved:
// the whole long is written but de-opt'ing will have to extract
// the relevant 32 bits.
const bool Matcher::int_in_long = true;
// Constants for c2c and c calling conventions.
const MachRegisterNumbers iarg_reg[8] = {
R3_num, R4_num, R5_num, R6_num,
R7_num, R8_num, R9_num, R10_num
};
const MachRegisterNumbers farg_reg[13] = {
F1_num, F2_num, F3_num, F4_num,
F5_num, F6_num, F7_num, F8_num,
F9_num, F10_num, F11_num, F12_num,
F13_num
};
const int num_iarg_registers = sizeof(iarg_reg) / sizeof(iarg_reg[0]);
const int num_farg_registers = sizeof(farg_reg) / sizeof(farg_reg[0]);
// Return whether or not this register is ever used as an argument. This
// function is used on startup to build the trampoline stubs in generateOptoStub.
// Registers not mentioned will be killed by the VM call in the trampoline, and
// arguments in those registers not be available to the callee.
bool Matcher::can_be_java_arg(int reg) {
// We return true for all registers contained in iarg_reg[] and
// farg_reg[] and their virtual halves.
// We must include the virtual halves in order to get STDs and LDs
// instead of STWs and LWs in the trampoline stubs.
if ( reg == R3_num || reg == R3_H_num
|| reg == R4_num || reg == R4_H_num
|| reg == R5_num || reg == R5_H_num
|| reg == R6_num || reg == R6_H_num
|| reg == R7_num || reg == R7_H_num
|| reg == R8_num || reg == R8_H_num
|| reg == R9_num || reg == R9_H_num
|| reg == R10_num || reg == R10_H_num)
return true;
if ( reg == F1_num || reg == F1_H_num
|| reg == F2_num || reg == F2_H_num
|| reg == F3_num || reg == F3_H_num
|| reg == F4_num || reg == F4_H_num
|| reg == F5_num || reg == F5_H_num
|| reg == F6_num || reg == F6_H_num
|| reg == F7_num || reg == F7_H_num
|| reg == F8_num || reg == F8_H_num
|| reg == F9_num || reg == F9_H_num
|| reg == F10_num || reg == F10_H_num
|| reg == F11_num || reg == F11_H_num
|| reg == F12_num || reg == F12_H_num
|| reg == F13_num || reg == F13_H_num)
return true;
return false;
}
bool Matcher::is_spillable_arg(int reg) {
return can_be_java_arg(reg);
}
bool Matcher::use_asm_for_ldiv_by_con(jlong divisor) {
return false;
}
// Register for DIVI projection of divmodI.
RegMask Matcher::divI_proj_mask() {
ShouldNotReachHere();
return RegMask();
}
// Register for MODI projection of divmodI.
RegMask Matcher::modI_proj_mask() {
ShouldNotReachHere();
return RegMask();
}
// Register for DIVL projection of divmodL.
RegMask Matcher::divL_proj_mask() {
ShouldNotReachHere();
return RegMask();
}
// Register for MODL projection of divmodL.
RegMask Matcher::modL_proj_mask() {
ShouldNotReachHere();
return RegMask();
}
const RegMask Matcher::method_handle_invoke_SP_save_mask() {
return RegMask();
}
%}
//----------ENCODING BLOCK-----------------------------------------------------
// This block specifies the encoding classes used by the compiler to output
// byte streams. Encoding classes are parameterized macros used by
// Machine Instruction Nodes in order to generate the bit encoding of the
// instruction. Operands specify their base encoding interface with the
// interface keyword. There are currently supported four interfaces,
// REG_INTER, CONST_INTER, MEMORY_INTER, & COND_INTER. REG_INTER causes an
// operand to generate a function which returns its register number when
// queried. CONST_INTER causes an operand to generate a function which
// returns the value of the constant when queried. MEMORY_INTER causes an
// operand to generate four functions which return the Base Register, the
// Index Register, the Scale Value, and the Offset Value of the operand when
// queried. COND_INTER causes an operand to generate six functions which
// return the encoding code (ie - encoding bits for the instruction)
// associated with each basic boolean condition for a conditional instruction.
//
// Instructions specify two basic values for encoding. Again, a function
// is available to check if the constant displacement is an oop. They use the
// ins_encode keyword to specify their encoding classes (which must be
// a sequence of enc_class names, and their parameters, specified in
// the encoding block), and they use the
// opcode keyword to specify, in order, their primary, secondary, and
// tertiary opcode. Only the opcode sections which a particular instruction
// needs for encoding need to be specified.
encode %{
enc_class enc_unimplemented %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
MacroAssembler _masm(&cbuf);
__ unimplemented("Unimplemented mach node encoding in AD file.", 13);
%}
enc_class enc_untested %{
#ifdef ASSERT
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
MacroAssembler _masm(&cbuf);
__ untested("Untested mach node encoding in AD file.");
#else
// TODO: PPC port $archOpcode(ppc64Opcode_none);
#endif
%}
enc_class enc_lbz(iRegIdst dst, memory mem) %{
// TODO: PPC port $archOpcode(ppc64Opcode_lbz);
MacroAssembler _masm(&cbuf);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
__ lbz($dst$$Register, Idisp, $mem$$base$$Register);
%}
// Load acquire.
enc_class enc_lbz_ac(iRegIdst dst, memory mem) %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
MacroAssembler _masm(&cbuf);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
__ lbz($dst$$Register, Idisp, $mem$$base$$Register);
__ twi_0($dst$$Register);
__ isync();
%}
enc_class enc_lhz(iRegIdst dst, memory mem) %{
// TODO: PPC port $archOpcode(ppc64Opcode_lhz);
MacroAssembler _masm(&cbuf);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
__ lhz($dst$$Register, Idisp, $mem$$base$$Register);
%}
// Load acquire.
enc_class enc_lhz_ac(iRegIdst dst, memory mem) %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
MacroAssembler _masm(&cbuf);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
__ lhz($dst$$Register, Idisp, $mem$$base$$Register);
__ twi_0($dst$$Register);
__ isync();
%}
enc_class enc_lwz(iRegIdst dst, memory mem) %{
// TODO: PPC port $archOpcode(ppc64Opcode_lwz);
MacroAssembler _masm(&cbuf);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
__ lwz($dst$$Register, Idisp, $mem$$base$$Register);
%}
// Load acquire.
enc_class enc_lwz_ac(iRegIdst dst, memory mem) %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
MacroAssembler _masm(&cbuf);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
__ lwz($dst$$Register, Idisp, $mem$$base$$Register);
__ twi_0($dst$$Register);
__ isync();
%}
enc_class enc_ld(iRegLdst dst, memoryAlg4 mem) %{
// TODO: PPC port $archOpcode(ppc64Opcode_ld);
MacroAssembler _masm(&cbuf);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
// Operand 'ds' requires 4-alignment.
assert((Idisp & 0x3) == 0, "unaligned offset");
__ ld($dst$$Register, Idisp, $mem$$base$$Register);
%}
// Load acquire.
enc_class enc_ld_ac(iRegLdst dst, memoryAlg4 mem) %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
MacroAssembler _masm(&cbuf);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
// Operand 'ds' requires 4-alignment.
assert((Idisp & 0x3) == 0, "unaligned offset");
__ ld($dst$$Register, Idisp, $mem$$base$$Register);
__ twi_0($dst$$Register);
__ isync();
%}
enc_class enc_lfd(RegF dst, memory mem) %{
// TODO: PPC port $archOpcode(ppc64Opcode_lfd);
MacroAssembler _masm(&cbuf);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
__ lfd($dst$$FloatRegister, Idisp, $mem$$base$$Register);
%}
enc_class enc_load_long_constL(iRegLdst dst, immL src, iRegLdst toc) %{
// TODO: PPC port $archOpcode(ppc64Opcode_ld);
MacroAssembler _masm(&cbuf);
int toc_offset = 0;
if (!ra_->C->in_scratch_emit_size()) {
address const_toc_addr;
// Create a non-oop constant, no relocation needed.
// If it is an IC, it has a virtual_call_Relocation.
const_toc_addr = __ long_constant((jlong)$src$$constant);
if (const_toc_addr == NULL) {
ciEnv::current()->record_out_of_memory_failure();
return;
}
// Get the constant's TOC offset.
toc_offset = __ offset_to_method_toc(const_toc_addr);
// Keep the current instruction offset in mind.
((loadConLNode*)this)->_cbuf_insts_offset = __ offset();
}
__ ld($dst$$Register, toc_offset, $toc$$Register);
%}
enc_class enc_load_long_constL_hi(iRegLdst dst, iRegLdst toc, immL src) %{
// TODO: PPC port $archOpcode(ppc64Opcode_addis);
MacroAssembler _masm(&cbuf);
if (!ra_->C->in_scratch_emit_size()) {
address const_toc_addr;
// Create a non-oop constant, no relocation needed.
// If it is an IC, it has a virtual_call_Relocation.
const_toc_addr = __ long_constant((jlong)$src$$constant);
if (const_toc_addr == NULL) {
ciEnv::current()->record_out_of_memory_failure();
return;
}
// Get the constant's TOC offset.
const int toc_offset = __ offset_to_method_toc(const_toc_addr);
// Store the toc offset of the constant.
((loadConL_hiNode*)this)->_const_toc_offset = toc_offset;
// Also keep the current instruction offset in mind.
((loadConL_hiNode*)this)->_cbuf_insts_offset = __ offset();
}
__ addis($dst$$Register, $toc$$Register, MacroAssembler::largeoffset_si16_si16_hi(_const_toc_offset));
%}
%} // encode
source %{
typedef struct {
loadConL_hiNode *_large_hi;
loadConL_loNode *_large_lo;
loadConLNode *_small;
MachNode *_last;
} loadConLNodesTuple;
loadConLNodesTuple loadConLNodesTuple_create(PhaseRegAlloc *ra_, Node *toc, immLOper *immSrc,
OptoReg::Name reg_second, OptoReg::Name reg_first) {
loadConLNodesTuple nodes;
const bool large_constant_pool = true; // TODO: PPC port C->cfg()->_consts_size > 4000;
if (large_constant_pool) {
// Create new nodes.
loadConL_hiNode *m1 = new loadConL_hiNode();
loadConL_loNode *m2 = new loadConL_loNode();
// inputs for new nodes
m1->add_req(NULL, toc);
m2->add_req(NULL, m1);
// operands for new nodes
m1->_opnds[0] = new iRegLdstOper(); // dst
m1->_opnds[1] = immSrc; // src
m1->_opnds[2] = new iRegPdstOper(); // toc
m2->_opnds[0] = new iRegLdstOper(); // dst
m2->_opnds[1] = immSrc; // src
m2->_opnds[2] = new iRegLdstOper(); // base
// Initialize ins_attrib TOC fields.
m1->_const_toc_offset = -1;
m2->_const_toc_offset_hi_node = m1;
// Initialize ins_attrib instruction offset.
m1->_cbuf_insts_offset = -1;
// register allocation for new nodes
ra_->set_pair(m1->_idx, reg_second, reg_first);
ra_->set_pair(m2->_idx, reg_second, reg_first);
// Create result.
nodes._large_hi = m1;
nodes._large_lo = m2;
nodes._small = NULL;
nodes._last = nodes._large_lo;
assert(m2->bottom_type()->isa_long(), "must be long");
} else {
loadConLNode *m2 = new loadConLNode();
// inputs for new nodes
m2->add_req(NULL, toc);
// operands for new nodes
m2->_opnds[0] = new iRegLdstOper(); // dst
m2->_opnds[1] = immSrc; // src
m2->_opnds[2] = new iRegPdstOper(); // toc
// Initialize ins_attrib instruction offset.
m2->_cbuf_insts_offset = -1;
// register allocation for new nodes
ra_->set_pair(m2->_idx, reg_second, reg_first);
// Create result.
nodes._large_hi = NULL;
nodes._large_lo = NULL;
nodes._small = m2;
nodes._last = nodes._small;
assert(m2->bottom_type()->isa_long(), "must be long");
}
return nodes;
}
%} // source
encode %{
// Postalloc expand emitter for loading a long constant from the method's TOC.
// Enc_class needed as consttanttablebase is not supported by postalloc
// expand.
enc_class postalloc_expand_load_long_constant(iRegLdst dst, immL src, iRegLdst toc) %{
// Create new nodes.
loadConLNodesTuple loadConLNodes =
loadConLNodesTuple_create(ra_, n_toc, op_src,
ra_->get_reg_second(this), ra_->get_reg_first(this));
// Push new nodes.
if (loadConLNodes._large_hi) nodes->push(loadConLNodes._large_hi);
if (loadConLNodes._last) nodes->push(loadConLNodes._last);
// some asserts
assert(nodes->length() >= 1, "must have created at least 1 node");
assert(loadConLNodes._last->bottom_type()->isa_long(), "must be long");
%}
enc_class enc_load_long_constP(iRegLdst dst, immP src, iRegLdst toc) %{
// TODO: PPC port $archOpcode(ppc64Opcode_ld);
MacroAssembler _masm(&cbuf);
int toc_offset = 0;
if (!ra_->C->in_scratch_emit_size()) {
intptr_t val = $src$$constant;
relocInfo::relocType constant_reloc = $src->constant_reloc(); // src
address const_toc_addr;
if (constant_reloc == relocInfo::oop_type) {
// Create an oop constant and a corresponding relocation.
AddressLiteral a = __ allocate_oop_address((jobject)val);
const_toc_addr = __ address_constant((address)a.value(), RelocationHolder::none);
__ relocate(a.rspec());
} else if (constant_reloc == relocInfo::metadata_type) {
AddressLiteral a = __ constant_metadata_address((Metadata *)val);
const_toc_addr = __ address_constant((address)a.value(), RelocationHolder::none);
__ relocate(a.rspec());
} else {
// Create a non-oop constant, no relocation needed.
const_toc_addr = __ long_constant((jlong)$src$$constant);
}
if (const_toc_addr == NULL) {
ciEnv::current()->record_out_of_memory_failure();
return;
}
// Get the constant's TOC offset.
toc_offset = __ offset_to_method_toc(const_toc_addr);
}
__ ld($dst$$Register, toc_offset, $toc$$Register);
%}
enc_class enc_load_long_constP_hi(iRegLdst dst, immP src, iRegLdst toc) %{
// TODO: PPC port $archOpcode(ppc64Opcode_addis);
MacroAssembler _masm(&cbuf);
if (!ra_->C->in_scratch_emit_size()) {
intptr_t val = $src$$constant;
relocInfo::relocType constant_reloc = $src->constant_reloc(); // src
address const_toc_addr;
if (constant_reloc == relocInfo::oop_type) {
// Create an oop constant and a corresponding relocation.
AddressLiteral a = __ allocate_oop_address((jobject)val);
const_toc_addr = __ address_constant((address)a.value(), RelocationHolder::none);
__ relocate(a.rspec());
} else if (constant_reloc == relocInfo::metadata_type) {
AddressLiteral a = __ constant_metadata_address((Metadata *)val);
const_toc_addr = __ address_constant((address)a.value(), RelocationHolder::none);
__ relocate(a.rspec());
} else { // non-oop pointers, e.g. card mark base, heap top
// Create a non-oop constant, no relocation needed.
const_toc_addr = __ long_constant((jlong)$src$$constant);
}
if (const_toc_addr == NULL) {
ciEnv::current()->record_out_of_memory_failure();
return;
}
// Get the constant's TOC offset.
const int toc_offset = __ offset_to_method_toc(const_toc_addr);
// Store the toc offset of the constant.
((loadConP_hiNode*)this)->_const_toc_offset = toc_offset;
}
__ addis($dst$$Register, $toc$$Register, MacroAssembler::largeoffset_si16_si16_hi(_const_toc_offset));
%}
// Postalloc expand emitter for loading a ptr constant from the method's TOC.
// Enc_class needed as consttanttablebase is not supported by postalloc
// expand.
enc_class postalloc_expand_load_ptr_constant(iRegPdst dst, immP src, iRegLdst toc) %{
const bool large_constant_pool = true; // TODO: PPC port C->cfg()->_consts_size > 4000;
if (large_constant_pool) {
// Create new nodes.
loadConP_hiNode *m1 = new loadConP_hiNode();
loadConP_loNode *m2 = new loadConP_loNode();
// inputs for new nodes
m1->add_req(NULL, n_toc);
m2->add_req(NULL, m1);
// operands for new nodes
m1->_opnds[0] = new iRegPdstOper(); // dst
m1->_opnds[1] = op_src; // src
m1->_opnds[2] = new iRegPdstOper(); // toc
m2->_opnds[0] = new iRegPdstOper(); // dst
m2->_opnds[1] = op_src; // src
m2->_opnds[2] = new iRegLdstOper(); // base
// Initialize ins_attrib TOC fields.
m1->_const_toc_offset = -1;
m2->_const_toc_offset_hi_node = m1;
// Register allocation for new nodes.
ra_->set_pair(m1->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
ra_->set_pair(m2->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
nodes->push(m1);
nodes->push(m2);
assert(m2->bottom_type()->isa_ptr(), "must be ptr");
} else {
loadConPNode *m2 = new loadConPNode();
// inputs for new nodes
m2->add_req(NULL, n_toc);
// operands for new nodes
m2->_opnds[0] = new iRegPdstOper(); // dst
m2->_opnds[1] = op_src; // src
m2->_opnds[2] = new iRegPdstOper(); // toc
// Register allocation for new nodes.
ra_->set_pair(m2->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
nodes->push(m2);
assert(m2->bottom_type()->isa_ptr(), "must be ptr");
}
%}
// Enc_class needed as consttanttablebase is not supported by postalloc
// expand.
enc_class postalloc_expand_load_float_constant(regF dst, immF src, iRegLdst toc) %{
bool large_constant_pool = true; // TODO: PPC port C->cfg()->_consts_size > 4000;
MachNode *m2;
if (large_constant_pool) {
m2 = new loadConFCompNode();
} else {
m2 = new loadConFNode();
}
// inputs for new nodes
m2->add_req(NULL, n_toc);
// operands for new nodes
m2->_opnds[0] = op_dst;
m2->_opnds[1] = op_src;
m2->_opnds[2] = new iRegPdstOper(); // constanttablebase
// register allocation for new nodes
ra_->set_pair(m2->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
nodes->push(m2);
%}
// Enc_class needed as consttanttablebase is not supported by postalloc
// expand.
enc_class postalloc_expand_load_double_constant(regD dst, immD src, iRegLdst toc) %{
bool large_constant_pool = true; // TODO: PPC port C->cfg()->_consts_size > 4000;
MachNode *m2;
if (large_constant_pool) {
m2 = new loadConDCompNode();
} else {
m2 = new loadConDNode();
}
// inputs for new nodes
m2->add_req(NULL, n_toc);
// operands for new nodes
m2->_opnds[0] = op_dst;
m2->_opnds[1] = op_src;
m2->_opnds[2] = new iRegPdstOper(); // constanttablebase
// register allocation for new nodes
ra_->set_pair(m2->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
nodes->push(m2);
%}
enc_class enc_stw(iRegIsrc src, memory mem) %{
// TODO: PPC port $archOpcode(ppc64Opcode_stw);
MacroAssembler _masm(&cbuf);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
__ stw($src$$Register, Idisp, $mem$$base$$Register);
%}
enc_class enc_std(iRegIsrc src, memoryAlg4 mem) %{
// TODO: PPC port $archOpcode(ppc64Opcode_std);
MacroAssembler _masm(&cbuf);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
// Operand 'ds' requires 4-alignment.
assert((Idisp & 0x3) == 0, "unaligned offset");
__ std($src$$Register, Idisp, $mem$$base$$Register);
%}
enc_class enc_stfs(RegF src, memory mem) %{
// TODO: PPC port $archOpcode(ppc64Opcode_stfs);
MacroAssembler _masm(&cbuf);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
__ stfs($src$$FloatRegister, Idisp, $mem$$base$$Register);
%}
enc_class enc_stfd(RegF src, memory mem) %{
// TODO: PPC port $archOpcode(ppc64Opcode_stfd);
MacroAssembler _masm(&cbuf);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
__ stfd($src$$FloatRegister, Idisp, $mem$$base$$Register);
%}
// Use release_store for card-marking to ensure that previous
// oop-stores are visible before the card-mark change.
enc_class enc_cms_card_mark(memory mem, iRegLdst releaseFieldAddr, flagsReg crx) %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
// FIXME: Implement this as a cmove and use a fixed condition code
// register which is written on every transition to compiled code,
// e.g. in call-stub and when returning from runtime stubs.
//
// Proposed code sequence for the cmove implementation:
//
// Label skip_release;
// __ beq(CCRfixed, skip_release);
// __ release();
// __ bind(skip_release);
// __ stb(card mark);
MacroAssembler _masm(&cbuf);
Label skip_storestore;
#if 0 // TODO: PPC port
// Check CMSCollectorCardTableModRefBSExt::_requires_release and do the
// StoreStore barrier conditionally.
__ lwz(R0, 0, $releaseFieldAddr$$Register);
__ cmpwi($crx$$CondRegister, R0, 0);
__ beq_predict_taken($crx$$CondRegister, skip_storestore);
#endif
__ li(R0, 0);
__ membar(Assembler::StoreStore);
#if 0 // TODO: PPC port
__ bind(skip_storestore);
#endif
// Do the store.
if ($mem$$index == 0) {
__ stb(R0, $mem$$disp, $mem$$base$$Register);
} else {
assert(0 == $mem$$disp, "no displacement possible with indexed load/stores on ppc");
__ stbx(R0, $mem$$base$$Register, $mem$$index$$Register);
}
%}
enc_class postalloc_expand_encode_oop(iRegNdst dst, iRegPdst src, flagsReg crx) %{
if (VM_Version::has_isel()) {
// use isel instruction with Power 7
cmpP_reg_imm16Node *n_compare = new cmpP_reg_imm16Node();
encodeP_subNode *n_sub_base = new encodeP_subNode();
encodeP_shiftNode *n_shift = new encodeP_shiftNode();
cond_set_0_oopNode *n_cond_set = new cond_set_0_oopNode();
n_compare->add_req(n_region, n_src);
n_compare->_opnds[0] = op_crx;
n_compare->_opnds[1] = op_src;
n_compare->_opnds[2] = new immL16Oper(0);
n_sub_base->add_req(n_region, n_src);
n_sub_base->_opnds[0] = op_dst;
n_sub_base->_opnds[1] = op_src;
n_sub_base->_bottom_type = _bottom_type;
n_shift->add_req(n_region, n_sub_base);
n_shift->_opnds[0] = op_dst;
n_shift->_opnds[1] = op_dst;
n_shift->_bottom_type = _bottom_type;
n_cond_set->add_req(n_region, n_compare, n_shift);
n_cond_set->_opnds[0] = op_dst;
n_cond_set->_opnds[1] = op_crx;
n_cond_set->_opnds[2] = op_dst;
n_cond_set->_bottom_type = _bottom_type;
ra_->set_pair(n_compare->_idx, ra_->get_reg_second(n_crx), ra_->get_reg_first(n_crx));
ra_->set_pair(n_sub_base->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
ra_->set_pair(n_shift->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
ra_->set_pair(n_cond_set->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
nodes->push(n_compare);
nodes->push(n_sub_base);
nodes->push(n_shift);
nodes->push(n_cond_set);
} else {
// before Power 7
moveRegNode *n_move = new moveRegNode();
cmpP_reg_imm16Node *n_compare = new cmpP_reg_imm16Node();
encodeP_shiftNode *n_shift = new encodeP_shiftNode();
cond_sub_baseNode *n_sub_base = new cond_sub_baseNode();
n_move->add_req(n_region, n_src);
n_move->_opnds[0] = op_dst;
n_move->_opnds[1] = op_src;
ra_->set_oop(n_move, true); // Until here, 'n_move' still produces an oop.
n_compare->add_req(n_region, n_src);
n_compare->add_prec(n_move);
n_compare->_opnds[0] = op_crx;
n_compare->_opnds[1] = op_src;
n_compare->_opnds[2] = new immL16Oper(0);
n_sub_base->add_req(n_region, n_compare, n_src);
n_sub_base->_opnds[0] = op_dst;
n_sub_base->_opnds[1] = op_crx;
n_sub_base->_opnds[2] = op_src;
n_sub_base->_bottom_type = _bottom_type;
n_shift->add_req(n_region, n_sub_base);
n_shift->_opnds[0] = op_dst;
n_shift->_opnds[1] = op_dst;
n_shift->_bottom_type = _bottom_type;
ra_->set_pair(n_shift->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
ra_->set_pair(n_compare->_idx, ra_->get_reg_second(n_crx), ra_->get_reg_first(n_crx));
ra_->set_pair(n_sub_base->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
ra_->set_pair(n_move->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
nodes->push(n_move);
nodes->push(n_compare);
nodes->push(n_sub_base);
nodes->push(n_shift);
}
assert(!(ra_->is_oop(this)), "sanity"); // This is not supposed to be GC'ed.
%}
enc_class postalloc_expand_encode_oop_not_null(iRegNdst dst, iRegPdst src) %{
encodeP_subNode *n1 = new encodeP_subNode();
n1->add_req(n_region, n_src);
n1->_opnds[0] = op_dst;
n1->_opnds[1] = op_src;
n1->_bottom_type = _bottom_type;
encodeP_shiftNode *n2 = new encodeP_shiftNode();
n2->add_req(n_region, n1);
n2->_opnds[0] = op_dst;
n2->_opnds[1] = op_dst;
n2->_bottom_type = _bottom_type;
ra_->set_pair(n1->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
ra_->set_pair(n2->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
nodes->push(n1);
nodes->push(n2);
assert(!(ra_->is_oop(this)), "sanity"); // This is not supposed to be GC'ed.
%}
enc_class postalloc_expand_decode_oop(iRegPdst dst, iRegNsrc src, flagsReg crx) %{
decodeN_shiftNode *n_shift = new decodeN_shiftNode();
cmpN_reg_imm0Node *n_compare = new cmpN_reg_imm0Node();
n_compare->add_req(n_region, n_src);
n_compare->_opnds[0] = op_crx;
n_compare->_opnds[1] = op_src;
n_compare->_opnds[2] = new immN_0Oper(TypeNarrowOop::NULL_PTR);
n_shift->add_req(n_region, n_src);
n_shift->_opnds[0] = op_dst;
n_shift->_opnds[1] = op_src;
n_shift->_bottom_type = _bottom_type;
if (VM_Version::has_isel()) {
// use isel instruction with Power 7
decodeN_addNode *n_add_base = new decodeN_addNode();
n_add_base->add_req(n_region, n_shift);
n_add_base->_opnds[0] = op_dst;
n_add_base->_opnds[1] = op_dst;
n_add_base->_bottom_type = _bottom_type;
cond_set_0_ptrNode *n_cond_set = new cond_set_0_ptrNode();
n_cond_set->add_req(n_region, n_compare, n_add_base);
n_cond_set->_opnds[0] = op_dst;
n_cond_set->_opnds[1] = op_crx;
n_cond_set->_opnds[2] = op_dst;
n_cond_set->_bottom_type = _bottom_type;
assert(ra_->is_oop(this) == true, "A decodeN node must produce an oop!");
ra_->set_oop(n_cond_set, true);
ra_->set_pair(n_shift->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
ra_->set_pair(n_compare->_idx, ra_->get_reg_second(n_crx), ra_->get_reg_first(n_crx));
ra_->set_pair(n_add_base->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
ra_->set_pair(n_cond_set->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
nodes->push(n_compare);
nodes->push(n_shift);
nodes->push(n_add_base);
nodes->push(n_cond_set);
} else {
// before Power 7
cond_add_baseNode *n_add_base = new cond_add_baseNode();
n_add_base->add_req(n_region, n_compare, n_shift);
n_add_base->_opnds[0] = op_dst;
n_add_base->_opnds[1] = op_crx;
n_add_base->_opnds[2] = op_dst;
n_add_base->_bottom_type = _bottom_type;
assert(ra_->is_oop(this) == true, "A decodeN node must produce an oop!");
ra_->set_oop(n_add_base, true);
ra_->set_pair(n_shift->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
ra_->set_pair(n_compare->_idx, ra_->get_reg_second(n_crx), ra_->get_reg_first(n_crx));
ra_->set_pair(n_add_base->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
nodes->push(n_compare);
nodes->push(n_shift);
nodes->push(n_add_base);
}
%}
enc_class postalloc_expand_decode_oop_not_null(iRegPdst dst, iRegNsrc src) %{
decodeN_shiftNode *n1 = new decodeN_shiftNode();
n1->add_req(n_region, n_src);
n1->_opnds[0] = op_dst;
n1->_opnds[1] = op_src;
n1->_bottom_type = _bottom_type;
decodeN_addNode *n2 = new decodeN_addNode();
n2->add_req(n_region, n1);
n2->_opnds[0] = op_dst;
n2->_opnds[1] = op_dst;
n2->_bottom_type = _bottom_type;
ra_->set_pair(n1->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
ra_->set_pair(n2->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
assert(ra_->is_oop(this) == true, "A decodeN node must produce an oop!");
ra_->set_oop(n2, true);
nodes->push(n1);
nodes->push(n2);
%}
enc_class enc_cmove_reg(iRegIdst dst, flagsRegSrc crx, iRegIsrc src, cmpOp cmp) %{
// TODO: PPC port $archOpcode(ppc64Opcode_cmove);
MacroAssembler _masm(&cbuf);
int cc = $cmp$$cmpcode;
int flags_reg = $crx$$reg;
Label done;
assert((Assembler::bcondCRbiIs1 & ~Assembler::bcondCRbiIs0) == 8, "check encoding");
// Branch if not (cmp crx).
__ bc(cc_to_inverse_boint(cc), cc_to_biint(cc, flags_reg), done);
__ mr($dst$$Register, $src$$Register);
// TODO PPC port __ endgroup_if_needed(_size == 12);
__ bind(done);
%}
enc_class enc_cmove_imm(iRegIdst dst, flagsRegSrc crx, immI16 src, cmpOp cmp) %{
// TODO: PPC port $archOpcode(ppc64Opcode_cmove);
MacroAssembler _masm(&cbuf);
Label done;
assert((Assembler::bcondCRbiIs1 & ~Assembler::bcondCRbiIs0) == 8, "check encoding");
// Branch if not (cmp crx).
__ bc(cc_to_inverse_boint($cmp$$cmpcode), cc_to_biint($cmp$$cmpcode, $crx$$reg), done);
__ li($dst$$Register, $src$$constant);
// TODO PPC port __ endgroup_if_needed(_size == 12);
__ bind(done);
%}
// New atomics.
enc_class enc_GetAndAddI(iRegIdst res, iRegPdst mem_ptr, iRegIsrc src) %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
MacroAssembler _masm(&cbuf);
Register Rtmp = R0;
Register Rres = $res$$Register;
Register Rsrc = $src$$Register;
Register Rptr = $mem_ptr$$Register;
bool RegCollision = (Rres == Rsrc) || (Rres == Rptr);
Register Rold = RegCollision ? Rtmp : Rres;
Label Lretry;
__ bind(Lretry);
__ lwarx(Rold, Rptr, MacroAssembler::cmpxchgx_hint_atomic_update());
__ add(Rtmp, Rsrc, Rold);
__ stwcx_(Rtmp, Rptr);
if (UseStaticBranchPredictionInCompareAndSwapPPC64) {
__ bne_predict_not_taken(CCR0, Lretry);
} else {
__ bne( CCR0, Lretry);
}
if (RegCollision) __ subf(Rres, Rsrc, Rtmp);
__ fence();
%}
enc_class enc_GetAndAddL(iRegLdst res, iRegPdst mem_ptr, iRegLsrc src) %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
MacroAssembler _masm(&cbuf);
Register Rtmp = R0;
Register Rres = $res$$Register;
Register Rsrc = $src$$Register;
Register Rptr = $mem_ptr$$Register;
bool RegCollision = (Rres == Rsrc) || (Rres == Rptr);
Register Rold = RegCollision ? Rtmp : Rres;
Label Lretry;
__ bind(Lretry);
__ ldarx(Rold, Rptr, MacroAssembler::cmpxchgx_hint_atomic_update());
__ add(Rtmp, Rsrc, Rold);
__ stdcx_(Rtmp, Rptr);
if (UseStaticBranchPredictionInCompareAndSwapPPC64) {
__ bne_predict_not_taken(CCR0, Lretry);
} else {
__ bne( CCR0, Lretry);
}
if (RegCollision) __ subf(Rres, Rsrc, Rtmp);
__ fence();
%}
enc_class enc_GetAndSetI(iRegIdst res, iRegPdst mem_ptr, iRegIsrc src) %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
MacroAssembler _masm(&cbuf);
Register Rtmp = R0;
Register Rres = $res$$Register;
Register Rsrc = $src$$Register;
Register Rptr = $mem_ptr$$Register;
bool RegCollision = (Rres == Rsrc) || (Rres == Rptr);
Register Rold = RegCollision ? Rtmp : Rres;
Label Lretry;
__ bind(Lretry);
__ lwarx(Rold, Rptr, MacroAssembler::cmpxchgx_hint_atomic_update());
__ stwcx_(Rsrc, Rptr);
if (UseStaticBranchPredictionInCompareAndSwapPPC64) {
__ bne_predict_not_taken(CCR0, Lretry);
} else {
__ bne( CCR0, Lretry);
}
if (RegCollision) __ mr(Rres, Rtmp);
__ fence();
%}
enc_class enc_GetAndSetL(iRegLdst res, iRegPdst mem_ptr, iRegLsrc src) %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
MacroAssembler _masm(&cbuf);
Register Rtmp = R0;
Register Rres = $res$$Register;
Register Rsrc = $src$$Register;
Register Rptr = $mem_ptr$$Register;
bool RegCollision = (Rres == Rsrc) || (Rres == Rptr);
Register Rold = RegCollision ? Rtmp : Rres;
Label Lretry;
__ bind(Lretry);
__ ldarx(Rold, Rptr, MacroAssembler::cmpxchgx_hint_atomic_update());
__ stdcx_(Rsrc, Rptr);
if (UseStaticBranchPredictionInCompareAndSwapPPC64) {
__ bne_predict_not_taken(CCR0, Lretry);
} else {
__ bne( CCR0, Lretry);
}
if (RegCollision) __ mr(Rres, Rtmp);
__ fence();
%}
// This enc_class is needed so that scheduler gets proper
// input mapping for latency computation.
enc_class enc_andc(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
// TODO: PPC port $archOpcode(ppc64Opcode_andc);
MacroAssembler _masm(&cbuf);
__ andc($dst$$Register, $src1$$Register, $src2$$Register);
%}
enc_class enc_convI2B_regI__cmove(iRegIdst dst, iRegIsrc src, flagsReg crx, immI16 zero, immI16 notzero) %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
MacroAssembler _masm(&cbuf);
Label done;
__ cmpwi($crx$$CondRegister, $src$$Register, 0);
__ li($dst$$Register, $zero$$constant);
__ beq($crx$$CondRegister, done);
__ li($dst$$Register, $notzero$$constant);
__ bind(done);
%}
enc_class enc_convP2B_regP__cmove(iRegIdst dst, iRegPsrc src, flagsReg crx, immI16 zero, immI16 notzero) %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
MacroAssembler _masm(&cbuf);
Label done;
__ cmpdi($crx$$CondRegister, $src$$Register, 0);
__ li($dst$$Register, $zero$$constant);
__ beq($crx$$CondRegister, done);
__ li($dst$$Register, $notzero$$constant);
__ bind(done);
%}
enc_class enc_cmove_bso_stackSlotL(iRegLdst dst, flagsRegSrc crx, stackSlotL mem ) %{
// TODO: PPC port $archOpcode(ppc64Opcode_cmove);
MacroAssembler _masm(&cbuf);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
Label done;
__ bso($crx$$CondRegister, done);
__ ld($dst$$Register, Idisp, $mem$$base$$Register);
// TODO PPC port __ endgroup_if_needed(_size == 12);
__ bind(done);
%}
enc_class enc_bc(flagsRegSrc crx, cmpOp cmp, Label lbl) %{
// TODO: PPC port $archOpcode(ppc64Opcode_bc);
MacroAssembler _masm(&cbuf);
Label d; // dummy
__ bind(d);
Label* p = ($lbl$$label);
// `p' is `NULL' when this encoding class is used only to
// determine the size of the encoded instruction.
Label& l = (NULL == p)? d : *(p);
int cc = $cmp$$cmpcode;
int flags_reg = $crx$$reg;
assert((Assembler::bcondCRbiIs1 & ~Assembler::bcondCRbiIs0) == 8, "check encoding");
int bhint = Assembler::bhintNoHint;
if (UseStaticBranchPredictionForUncommonPathsPPC64) {
if (_prob <= PROB_NEVER) {
bhint = Assembler::bhintIsNotTaken;
} else if (_prob >= PROB_ALWAYS) {
bhint = Assembler::bhintIsTaken;
}
}
__ bc(Assembler::add_bhint_to_boint(bhint, cc_to_boint(cc)),
cc_to_biint(cc, flags_reg),
l);
%}
enc_class enc_bc_far(flagsRegSrc crx, cmpOp cmp, Label lbl) %{
// The scheduler doesn't know about branch shortening, so we set the opcode
// to ppc64Opcode_bc in order to hide this detail from the scheduler.
// TODO: PPC port $archOpcode(ppc64Opcode_bc);
MacroAssembler _masm(&cbuf);
Label d; // dummy
__ bind(d);
Label* p = ($lbl$$label);
// `p' is `NULL' when this encoding class is used only to
// determine the size of the encoded instruction.
Label& l = (NULL == p)? d : *(p);
int cc = $cmp$$cmpcode;
int flags_reg = $crx$$reg;
int bhint = Assembler::bhintNoHint;
if (UseStaticBranchPredictionForUncommonPathsPPC64) {
if (_prob <= PROB_NEVER) {
bhint = Assembler::bhintIsNotTaken;
} else if (_prob >= PROB_ALWAYS) {
bhint = Assembler::bhintIsTaken;
}
}
// Tell the conditional far branch to optimize itself when being relocated.
__ bc_far(Assembler::add_bhint_to_boint(bhint, cc_to_boint(cc)),
cc_to_biint(cc, flags_reg),
l,
MacroAssembler::bc_far_optimize_on_relocate);
%}
// Branch used with Power6 scheduling (can be shortened without changing the node).
enc_class enc_bc_short_far(flagsRegSrc crx, cmpOp cmp, Label lbl) %{
// The scheduler doesn't know about branch shortening, so we set the opcode
// to ppc64Opcode_bc in order to hide this detail from the scheduler.
// TODO: PPC port $archOpcode(ppc64Opcode_bc);
MacroAssembler _masm(&cbuf);
Label d; // dummy
__ bind(d);
Label* p = ($lbl$$label);
// `p' is `NULL' when this encoding class is used only to
// determine the size of the encoded instruction.
Label& l = (NULL == p)? d : *(p);
int cc = $cmp$$cmpcode;
int flags_reg = $crx$$reg;
int bhint = Assembler::bhintNoHint;
if (UseStaticBranchPredictionForUncommonPathsPPC64) {
if (_prob <= PROB_NEVER) {
bhint = Assembler::bhintIsNotTaken;
} else if (_prob >= PROB_ALWAYS) {
bhint = Assembler::bhintIsTaken;
}
}
#if 0 // TODO: PPC port
if (_size == 8) {
// Tell the conditional far branch to optimize itself when being relocated.
__ bc_far(Assembler::add_bhint_to_boint(bhint, cc_to_boint(cc)),
cc_to_biint(cc, flags_reg),
l,
MacroAssembler::bc_far_optimize_on_relocate);
} else {
__ bc (Assembler::add_bhint_to_boint(bhint, cc_to_boint(cc)),
cc_to_biint(cc, flags_reg),
l);
}
#endif
Unimplemented();
%}
// Postalloc expand emitter for loading a replicatef float constant from
// the method's TOC.
// Enc_class needed as consttanttablebase is not supported by postalloc
// expand.
enc_class postalloc_expand_load_replF_constant(iRegLdst dst, immF src, iRegLdst toc) %{
// Create new nodes.
// Make an operand with the bit pattern to load as float.
immLOper *op_repl = new immLOper((jlong)replicate_immF(op_src->constantF()));
loadConLNodesTuple loadConLNodes =
loadConLNodesTuple_create(ra_, n_toc, op_repl,
ra_->get_reg_second(this), ra_->get_reg_first(this));
// Push new nodes.
if (loadConLNodes._large_hi) nodes->push(loadConLNodes._large_hi);
if (loadConLNodes._last) nodes->push(loadConLNodes._last);
assert(nodes->length() >= 1, "must have created at least 1 node");
assert(loadConLNodes._last->bottom_type()->isa_long(), "must be long");
%}
// This enc_class is needed so that scheduler gets proper
// input mapping for latency computation.
enc_class enc_poll(immI dst, iRegLdst poll) %{
// TODO: PPC port $archOpcode(ppc64Opcode_ld);
// Fake operand dst needed for PPC scheduler.
assert($dst$$constant == 0x0, "dst must be 0x0");
MacroAssembler _masm(&cbuf);
// Mark the code position where the load from the safepoint
// polling page was emitted as relocInfo::poll_type.
__ relocate(relocInfo::poll_type);
__ load_from_polling_page($poll$$Register);
%}
// A Java static call or a runtime call.
//
// Branch-and-link relative to a trampoline.
// The trampoline loads the target address and does a long branch to there.
// In case we call java, the trampoline branches to a interpreter_stub
// which loads the inline cache and the real call target from the constant pool.
//
// This basically looks like this:
//
// >>>> consts -+ -+
// | |- offset1
// [call target1] | <-+
// [IC cache] |- offset2
// [call target2] <--+
//
// <<<< consts
// >>>> insts
//
// bl offset16 -+ -+ ??? // How many bits available?
// | |
// <<<< insts | |
// >>>> stubs | |
// | |- trampoline_stub_Reloc
// trampoline stub: | <-+
// r2 = toc |
// r2 = [r2 + offset1] | // Load call target1 from const section
// mtctr r2 |
// bctr |- static_stub_Reloc
// comp_to_interp_stub: <---+
// r1 = toc
// ICreg = [r1 + IC_offset] // Load IC from const section
// r1 = [r1 + offset2] // Load call target2 from const section
// mtctr r1
// bctr
//
// <<<< stubs
//
// The call instruction in the code either
// - Branches directly to a compiled method if the offset is encodable in instruction.
// - Branches to the trampoline stub if the offset to the compiled method is not encodable.
// - Branches to the compiled_to_interp stub if the target is interpreted.
//
// Further there are three relocations from the loads to the constants in
// the constant section.
//
// Usage of r1 and r2 in the stubs allows to distinguish them.
enc_class enc_java_static_call(method meth) %{
// TODO: PPC port $archOpcode(ppc64Opcode_bl);
MacroAssembler _masm(&cbuf);
address entry_point = (address)$meth$$method;
if (!_method) {
// A call to a runtime wrapper, e.g. new, new_typeArray_Java, uncommon_trap.
emit_call_with_trampoline_stub(_masm, entry_point, relocInfo::runtime_call_type);
} else {
// Remember the offset not the address.
const int start_offset = __ offset();
// The trampoline stub.
if (!Compile::current()->in_scratch_emit_size()) {
// No entry point given, use the current pc.
// Make sure branch fits into
if (entry_point == 0) entry_point = __ pc();
// Put the entry point as a constant into the constant pool.
const address entry_point_toc_addr = __ address_constant(entry_point, RelocationHolder::none);
if (entry_point_toc_addr == NULL) {
ciEnv::current()->record_out_of_memory_failure();
return;
}
const int entry_point_toc_offset = __ offset_to_method_toc(entry_point_toc_addr);
// Emit the trampoline stub which will be related to the branch-and-link below.
CallStubImpl::emit_trampoline_stub(_masm, entry_point_toc_offset, start_offset);
if (ciEnv::current()->failing()) { return; } // Code cache may be full.
int method_index = resolved_method_index(cbuf);
__ relocate(_optimized_virtual ? opt_virtual_call_Relocation::spec(method_index)
: static_call_Relocation::spec(method_index));
}
// The real call.
// Note: At this point we do not have the address of the trampoline
// stub, and the entry point might be too far away for bl, so __ pc()
// serves as dummy and the bl will be patched later.
cbuf.set_insts_mark();
__ bl(__ pc()); // Emits a relocation.
// The stub for call to interpreter.
address stub = CompiledStaticCall::emit_to_interp_stub(cbuf);
if (stub == NULL) {
ciEnv::current()->record_failure("CodeCache is full");
return;
}
}
%}
// Second node of expanded dynamic call - the call.
enc_class enc_java_dynamic_call_sched(method meth) %{
// TODO: PPC port $archOpcode(ppc64Opcode_bl);
MacroAssembler _masm(&cbuf);
if (!ra_->C->in_scratch_emit_size()) {
// Create a call trampoline stub for the given method.
const address entry_point = !($meth$$method) ? 0 : (address)$meth$$method;
const address entry_point_const = __ address_constant(entry_point, RelocationHolder::none);
if (entry_point_const == NULL) {
ciEnv::current()->record_out_of_memory_failure();
return;
}
const int entry_point_const_toc_offset = __ offset_to_method_toc(entry_point_const);
CallStubImpl::emit_trampoline_stub(_masm, entry_point_const_toc_offset, __ offset());
if (ra_->C->env()->failing()) { return; } // Code cache may be full.
// Build relocation at call site with ic position as data.
assert((_load_ic_hi_node != NULL && _load_ic_node == NULL) ||
(_load_ic_hi_node == NULL && _load_ic_node != NULL),
"must have one, but can't have both");
assert((_load_ic_hi_node != NULL && _load_ic_hi_node->_cbuf_insts_offset != -1) ||
(_load_ic_node != NULL && _load_ic_node->_cbuf_insts_offset != -1),
"must contain instruction offset");
const int virtual_call_oop_addr_offset = _load_ic_hi_node != NULL
? _load_ic_hi_node->_cbuf_insts_offset
: _load_ic_node->_cbuf_insts_offset;
const address virtual_call_oop_addr = __ addr_at(virtual_call_oop_addr_offset);
assert(MacroAssembler::is_load_const_from_method_toc_at(virtual_call_oop_addr),
"should be load from TOC");
int method_index = resolved_method_index(cbuf);
__ relocate(virtual_call_Relocation::spec(virtual_call_oop_addr, method_index));
}
// At this point I do not have the address of the trampoline stub,
// and the entry point might be too far away for bl. Pc() serves
// as dummy and bl will be patched later.
__ bl((address) __ pc());
%}
// postalloc expand emitter for virtual calls.
enc_class postalloc_expand_java_dynamic_call_sched(method meth, iRegLdst toc) %{
// Create the nodes for loading the IC from the TOC.
loadConLNodesTuple loadConLNodes_IC =
loadConLNodesTuple_create(ra_, n_toc, new immLOper((jlong)Universe::non_oop_word()),
OptoReg::Name(R19_H_num), OptoReg::Name(R19_num));
// Create the call node.
CallDynamicJavaDirectSchedNode *call = new CallDynamicJavaDirectSchedNode();
call->_method_handle_invoke = _method_handle_invoke;
call->_vtable_index = _vtable_index;
call->_method = _method;
call->_bci = _bci;
call->_optimized_virtual = _optimized_virtual;
call->_tf = _tf;
call->_entry_point = _entry_point;
call->_cnt = _cnt;
call->_argsize = _argsize;
call->_oop_map = _oop_map;
call->_jvms = _jvms;
call->_jvmadj = _jvmadj;
call->_in_rms = _in_rms;
call->_nesting = _nesting;
call->_override_symbolic_info = _override_symbolic_info;
// New call needs all inputs of old call.
// Req...
for (uint i = 0; i < req(); ++i) {
// The expanded node does not need toc any more.
// Add the inline cache constant here instead. This expresses the
// register of the inline cache must be live at the call.
// Else we would have to adapt JVMState by -1.
if (i == mach_constant_base_node_input()) {
call->add_req(loadConLNodes_IC._last);
} else {
call->add_req(in(i));
}
}
// ...as well as prec
for (uint i = req(); i < len(); ++i) {
call->add_prec(in(i));
}
// Remember nodes loading the inline cache into r19.
call->_load_ic_hi_node = loadConLNodes_IC._large_hi;
call->_load_ic_node = loadConLNodes_IC._small;
// Operands for new nodes.
call->_opnds[0] = _opnds[0];
call->_opnds[1] = _opnds[1];
// Only the inline cache is associated with a register.
assert(Matcher::inline_cache_reg() == OptoReg::Name(R19_num), "ic reg should be R19");
// Push new nodes.
if (loadConLNodes_IC._large_hi) nodes->push(loadConLNodes_IC._large_hi);
if (loadConLNodes_IC._last) nodes->push(loadConLNodes_IC._last);
nodes->push(call);
%}
// Compound version of call dynamic
// Toc is only passed so that it can be used in ins_encode statement.
// In the code we have to use $constanttablebase.
enc_class enc_java_dynamic_call(method meth, iRegLdst toc) %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
MacroAssembler _masm(&cbuf);
int start_offset = __ offset();
Register Rtoc = (ra_) ? $constanttablebase : R2_TOC;
#if 0
int vtable_index = this->_vtable_index;
if (_vtable_index < 0) {
// Must be invalid_vtable_index, not nonvirtual_vtable_index.
assert(_vtable_index == Method::invalid_vtable_index, "correct sentinel value");
Register ic_reg = as_Register(Matcher::inline_cache_reg_encode());
// Virtual call relocation will point to ic load.
address virtual_call_meta_addr = __ pc();
// Load a clear inline cache.
AddressLiteral empty_ic((address) Universe::non_oop_word());
bool success = __ load_const_from_method_toc(ic_reg, empty_ic, Rtoc, /*fixed_size*/ true);
if (!success) {
ciEnv::current()->record_out_of_memory_failure();
return;
}
// CALL to fixup routine. Fixup routine uses ScopeDesc info
// to determine who we intended to call.
__ relocate(virtual_call_Relocation::spec(virtual_call_meta_addr));
emit_call_with_trampoline_stub(_masm, (address)$meth$$method, relocInfo::none);
assert(((MachCallDynamicJavaNode*)this)->ret_addr_offset() == __ offset() - start_offset,
"Fix constant in ret_addr_offset()");
} else {
assert(!UseInlineCaches, "expect vtable calls only if not using ICs");
// Go thru the vtable. Get receiver klass. Receiver already
// checked for non-null. If we'll go thru a C2I adapter, the
// interpreter expects method in R19_method.
__ load_klass(R11_scratch1, R3);
int entry_offset = in_bytes(Klass::vtable_start_offset()) + _vtable_index * vtableEntry::size_in_bytes();
int v_off = entry_offset + vtableEntry::method_offset_in_bytes();
__ li(R19_method, v_off);
__ ldx(R19_method/*method oop*/, R19_method/*method offset*/, R11_scratch1/*class*/);
// NOTE: for vtable dispatches, the vtable entry will never be
// null. However it may very well end up in handle_wrong_method
// if the method is abstract for the particular class.
__ ld(R11_scratch1, in_bytes(Method::from_compiled_offset()), R19_method);
// Call target. Either compiled code or C2I adapter.
__ mtctr(R11_scratch1);
__ bctrl();
if (((MachCallDynamicJavaNode*)this)->ret_addr_offset() != __ offset() - start_offset) {
tty->print(" %d, %d\n", ((MachCallDynamicJavaNode*)this)->ret_addr_offset(),__ offset() - start_offset);
}
assert(((MachCallDynamicJavaNode*)this)->ret_addr_offset() == __ offset() - start_offset,
"Fix constant in ret_addr_offset()");
}
#endif
Unimplemented(); // ret_addr_offset not yet fixed. Depends on compressed oops (load klass!).
%}
// a runtime call
enc_class enc_java_to_runtime_call (method meth) %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
MacroAssembler _masm(&cbuf);
const address start_pc = __ pc();
#if defined(ABI_ELFv2)
address entry= !($meth$$method) ? NULL : (address)$meth$$method;
__ call_c(entry, relocInfo::runtime_call_type);
#else
// The function we're going to call.
FunctionDescriptor fdtemp;
const FunctionDescriptor* fd = !($meth$$method) ? &fdtemp : (FunctionDescriptor*)$meth$$method;
Register Rtoc = R12_scratch2;
// Calculate the method's TOC.
__ calculate_address_from_global_toc(Rtoc, __ method_toc());
// Put entry, env, toc into the constant pool, this needs up to 3 constant
// pool entries; call_c_using_toc will optimize the call.
bool success = __ call_c_using_toc(fd, relocInfo::runtime_call_type, Rtoc);
if (!success) {
ciEnv::current()->record_out_of_memory_failure();
return;
}
#endif
// Check the ret_addr_offset.
assert(((MachCallRuntimeNode*)this)->ret_addr_offset() == __ last_calls_return_pc() - start_pc,
"Fix constant in ret_addr_offset()");
%}
// Move to ctr for leaf call.
// This enc_class is needed so that scheduler gets proper
// input mapping for latency computation.
enc_class enc_leaf_call_mtctr(iRegLsrc src) %{
// TODO: PPC port $archOpcode(ppc64Opcode_mtctr);
MacroAssembler _masm(&cbuf);
__ mtctr($src$$Register);
%}
// Postalloc expand emitter for runtime leaf calls.
enc_class postalloc_expand_java_to_runtime_call(method meth, iRegLdst toc) %{
loadConLNodesTuple loadConLNodes_Entry;
#if defined(ABI_ELFv2)
jlong entry_address = (jlong) this->entry_point();
assert(entry_address, "need address here");
loadConLNodes_Entry = loadConLNodesTuple_create(ra_, n_toc, new immLOper(entry_address),
OptoReg::Name(R12_H_num), OptoReg::Name(R12_num));
#else
// Get the struct that describes the function we are about to call.
FunctionDescriptor* fd = (FunctionDescriptor*) this->entry_point();
assert(fd, "need fd here");
jlong entry_address = (jlong) fd->entry();
// new nodes
loadConLNodesTuple loadConLNodes_Env;
loadConLNodesTuple loadConLNodes_Toc;
// Create nodes and operands for loading the entry point.
loadConLNodes_Entry = loadConLNodesTuple_create(ra_, n_toc, new immLOper(entry_address),
OptoReg::Name(R12_H_num), OptoReg::Name(R12_num));
// Create nodes and operands for loading the env pointer.
if (fd->env() != NULL) {
loadConLNodes_Env = loadConLNodesTuple_create(ra_, n_toc, new immLOper((jlong) fd->env()),
OptoReg::Name(R11_H_num), OptoReg::Name(R11_num));
} else {
loadConLNodes_Env._large_hi = NULL;
loadConLNodes_Env._large_lo = NULL;
loadConLNodes_Env._small = NULL;
loadConLNodes_Env._last = new loadConL16Node();
loadConLNodes_Env._last->_opnds[0] = new iRegLdstOper();
loadConLNodes_Env._last->_opnds[1] = new immL16Oper(0);
ra_->set_pair(loadConLNodes_Env._last->_idx, OptoReg::Name(R11_H_num), OptoReg::Name(R11_num));
}
// Create nodes and operands for loading the Toc point.
loadConLNodes_Toc = loadConLNodesTuple_create(ra_, n_toc, new immLOper((jlong) fd->toc()),
OptoReg::Name(R2_H_num), OptoReg::Name(R2_num));
#endif // ABI_ELFv2
// mtctr node
MachNode *mtctr = new CallLeafDirect_mtctrNode();
assert(loadConLNodes_Entry._last != NULL, "entry must exist");
mtctr->add_req(0, loadConLNodes_Entry._last);
mtctr->_opnds[0] = new iRegLdstOper();
mtctr->_opnds[1] = new iRegLdstOper();
// call node
MachCallLeafNode *call = new CallLeafDirectNode();
call->_opnds[0] = _opnds[0];
call->_opnds[1] = new methodOper((intptr_t) entry_address); // May get set later.
// Make the new call node look like the old one.
call->_name = _name;
call->_tf = _tf;
call->_entry_point = _entry_point;
call->_cnt = _cnt;
call->_argsize = _argsize;
call->_oop_map = _oop_map;
guarantee(!_jvms, "You must clone the jvms and adapt the offsets by fix_jvms().");
call->_jvms = NULL;
call->_jvmadj = _jvmadj;
call->_in_rms = _in_rms;
call->_nesting = _nesting;
// New call needs all inputs of old call.
// Req...
for (uint i = 0; i < req(); ++i) {
if (i != mach_constant_base_node_input()) {
call->add_req(in(i));
}
}
// These must be reqired edges, as the registers are live up to
// the call. Else the constants are handled as kills.
call->add_req(mtctr);
#if !defined(ABI_ELFv2)
call->add_req(loadConLNodes_Env._last);
call->add_req(loadConLNodes_Toc._last);
#endif
// ...as well as prec
for (uint i = req(); i < len(); ++i) {
call->add_prec(in(i));
}
// registers
ra_->set1(mtctr->_idx, OptoReg::Name(SR_CTR_num));
// Insert the new nodes.
if (loadConLNodes_Entry._large_hi) nodes->push(loadConLNodes_Entry._large_hi);
if (loadConLNodes_Entry._last) nodes->push(loadConLNodes_Entry._last);
#if !defined(ABI_ELFv2)
if (loadConLNodes_Env._large_hi) nodes->push(loadConLNodes_Env._large_hi);
if (loadConLNodes_Env._last) nodes->push(loadConLNodes_Env._last);
if (loadConLNodes_Toc._large_hi) nodes->push(loadConLNodes_Toc._large_hi);
if (loadConLNodes_Toc._last) nodes->push(loadConLNodes_Toc._last);
#endif
nodes->push(mtctr);
nodes->push(call);
%}
%}
//----------FRAME--------------------------------------------------------------
// Definition of frame structure and management information.
frame %{
// What direction does stack grow in (assumed to be same for native & Java).
stack_direction(TOWARDS_LOW);
// These two registers define part of the calling convention between
// compiled code and the interpreter.
// Inline Cache Register or method for I2C.
inline_cache_reg(R19); // R19_method
// Method Oop Register when calling interpreter.
interpreter_method_oop_reg(R19); // R19_method
// Optional: name the operand used by cisc-spilling to access
// [stack_pointer + offset].
cisc_spilling_operand_name(indOffset);
// Number of stack slots consumed by a Monitor enter.
sync_stack_slots((frame::jit_monitor_size / VMRegImpl::stack_slot_size));
// Compiled code's Frame Pointer.
frame_pointer(R1); // R1_SP
// Interpreter stores its frame pointer in a register which is
// stored to the stack by I2CAdaptors. I2CAdaptors convert from
// interpreted java to compiled java.
//
// R14_state holds pointer to caller's cInterpreter.
interpreter_frame_pointer(R14); // R14_state
stack_alignment(frame::alignment_in_bytes);
in_preserve_stack_slots((frame::jit_in_preserve_size / VMRegImpl::stack_slot_size));
// Number of outgoing stack slots killed above the
// out_preserve_stack_slots for calls to C. Supports the var-args
// backing area for register parms.
//
varargs_C_out_slots_killed(((frame::abi_reg_args_size - frame::jit_out_preserve_size) / VMRegImpl::stack_slot_size));
// The after-PROLOG location of the return address. Location of
// return address specifies a type (REG or STACK) and a number
// representing the register number (i.e. - use a register name) or
// stack slot.
//
// A: Link register is stored in stack slot ...
// M: ... but it's in the caller's frame according to PPC-64 ABI.
// J: Therefore, we make sure that the link register is also in R11_scratch1
// at the end of the prolog.
// B: We use R20, now.
//return_addr(REG R20);
// G: After reading the comments made by all the luminaries on their
// failure to tell the compiler where the return address really is,
// I hardly dare to try myself. However, I'm convinced it's in slot
// 4 what apparently works and saves us some spills.
return_addr(STACK 4);
// This is the body of the function
//
// void Matcher::calling_convention(OptoRegPair* sig, // array of ideal regs
// uint length, // length of array
// bool is_outgoing)
//
// The `sig' array is to be updated. sig[j] represents the location
// of the j-th argument, either a register or a stack slot.
// Comment taken from i486.ad:
// Body of function which returns an integer array locating
// arguments either in registers or in stack slots. Passed an array
// of ideal registers called "sig" and a "length" count. Stack-slot
// offsets are based on outgoing arguments, i.e. a CALLER setting up
// arguments for a CALLEE. Incoming stack arguments are
// automatically biased by the preserve_stack_slots field above.
calling_convention %{
// No difference between ingoing/outgoing. Just pass false.
SharedRuntime::java_calling_convention(sig_bt, regs, length, false);
%}
// Comment taken from i486.ad:
// Body of function which returns an integer array locating
// arguments either in registers or in stack slots. Passed an array
// of ideal registers called "sig" and a "length" count. Stack-slot
// offsets are based on outgoing arguments, i.e. a CALLER setting up
// arguments for a CALLEE. Incoming stack arguments are
// automatically biased by the preserve_stack_slots field above.
c_calling_convention %{
// This is obviously always outgoing.
// C argument in register AND stack slot.
(void) SharedRuntime::c_calling_convention(sig_bt, regs, /*regs2=*/NULL, length);
%}
// Location of native (C/C++) and interpreter return values. This
// is specified to be the same as Java. In the 32-bit VM, long
// values are actually returned from native calls in O0:O1 and
// returned to the interpreter in I0:I1. The copying to and from
// the register pairs is done by the appropriate call and epilog
// opcodes. This simplifies the register allocator.
c_return_value %{
assert((ideal_reg >= Op_RegI && ideal_reg <= Op_RegL) ||
(ideal_reg == Op_RegN && Universe::narrow_oop_base() == NULL && Universe::narrow_oop_shift() == 0),
"only return normal values");
// enum names from opcodes.hpp: Op_Node Op_Set Op_RegN Op_RegI Op_RegP Op_RegF Op_RegD Op_RegL
static int typeToRegLo[Op_RegL+1] = { 0, 0, R3_num, R3_num, R3_num, F1_num, F1_num, R3_num };
static int typeToRegHi[Op_RegL+1] = { 0, 0, OptoReg::Bad, R3_H_num, R3_H_num, OptoReg::Bad, F1_H_num, R3_H_num };
return OptoRegPair(typeToRegHi[ideal_reg], typeToRegLo[ideal_reg]);
%}
// Location of compiled Java return values. Same as C
return_value %{
assert((ideal_reg >= Op_RegI && ideal_reg <= Op_RegL) ||
(ideal_reg == Op_RegN && Universe::narrow_oop_base() == NULL && Universe::narrow_oop_shift() == 0),
"only return normal values");
// enum names from opcodes.hpp: Op_Node Op_Set Op_RegN Op_RegI Op_RegP Op_RegF Op_RegD Op_RegL
static int typeToRegLo[Op_RegL+1] = { 0, 0, R3_num, R3_num, R3_num, F1_num, F1_num, R3_num };
static int typeToRegHi[Op_RegL+1] = { 0, 0, OptoReg::Bad, R3_H_num, R3_H_num, OptoReg::Bad, F1_H_num, R3_H_num };
return OptoRegPair(typeToRegHi[ideal_reg], typeToRegLo[ideal_reg]);
%}
%}
//----------ATTRIBUTES---------------------------------------------------------
//----------Operand Attributes-------------------------------------------------
op_attrib op_cost(1); // Required cost attribute.
//----------Instruction Attributes---------------------------------------------
// Cost attribute. required.
ins_attrib ins_cost(DEFAULT_COST);
// Is this instruction a non-matching short branch variant of some
// long branch? Not required.
ins_attrib ins_short_branch(0);
ins_attrib ins_is_TrapBasedCheckNode(true);
// Number of constants.
// This instruction uses the given number of constants
// (optional attribute).
// This is needed to determine in time whether the constant pool will
// exceed 4000 entries. Before postalloc_expand the overall number of constants
// is determined. It's also used to compute the constant pool size
// in Output().
ins_attrib ins_num_consts(0);
// Required alignment attribute (must be a power of 2) specifies the
// alignment that some part of the instruction (not necessarily the
// start) requires. If > 1, a compute_padding() function must be
// provided for the instruction.
ins_attrib ins_alignment(1);
// Enforce/prohibit rematerializations.
// - If an instruction is attributed with 'ins_cannot_rematerialize(true)'
// then rematerialization of that instruction is prohibited and the
// instruction's value will be spilled if necessary.
// Causes that MachNode::rematerialize() returns false.
// - If an instruction is attributed with 'ins_should_rematerialize(true)'
// then rematerialization should be enforced and a copy of the instruction
// should be inserted if possible; rematerialization is not guaranteed.
// Note: this may result in rematerializations in front of every use.
// Causes that MachNode::rematerialize() can return true.
// (optional attribute)
ins_attrib ins_cannot_rematerialize(false);
ins_attrib ins_should_rematerialize(false);
// Instruction has variable size depending on alignment.
ins_attrib ins_variable_size_depending_on_alignment(false);
// Instruction is a nop.
ins_attrib ins_is_nop(false);
// Instruction is mapped to a MachIfFastLock node (instead of MachFastLock).
ins_attrib ins_use_mach_if_fast_lock_node(false);
// Field for the toc offset of a constant.
//
// This is needed if the toc offset is not encodable as an immediate in
// the PPC load instruction. If so, the upper (hi) bits of the offset are
// added to the toc, and from this a load with immediate is performed.
// With postalloc expand, we get two nodes that require the same offset
// but which don't know about each other. The offset is only known
// when the constant is added to the constant pool during emitting.
// It is generated in the 'hi'-node adding the upper bits, and saved
// in this node. The 'lo'-node has a link to the 'hi'-node and reads
// the offset from there when it gets encoded.
ins_attrib ins_field_const_toc_offset(0);
ins_attrib ins_field_const_toc_offset_hi_node(0);
// A field that can hold the instructions offset in the code buffer.
// Set in the nodes emitter.
ins_attrib ins_field_cbuf_insts_offset(-1);
// Fields for referencing a call's load-IC-node.
// If the toc offset can not be encoded as an immediate in a load, we
// use two nodes.
ins_attrib ins_field_load_ic_hi_node(0);
ins_attrib ins_field_load_ic_node(0);
//----------OPERANDS-----------------------------------------------------------
// Operand definitions must precede instruction definitions for correct
// parsing in the ADLC because operands constitute user defined types
// which are used in instruction definitions.
//
// Formats are generated automatically for constants and base registers.
//----------Simple Operands----------------------------------------------------
// Immediate Operands
// Integer Immediate: 32-bit
operand immI() %{
match(ConI);
op_cost(40);
format %{ %}
interface(CONST_INTER);
%}
operand immI8() %{
predicate(Assembler::is_simm(n->get_int(), 8));
op_cost(0);
match(ConI);
format %{ %}
interface(CONST_INTER);
%}
// Integer Immediate: 16-bit
operand immI16() %{
predicate(Assembler::is_simm(n->get_int(), 16));
op_cost(0);
match(ConI);
format %{ %}
interface(CONST_INTER);
%}
// Integer Immediate: 32-bit, where lowest 16 bits are 0x0000.
operand immIhi16() %{
predicate(((n->get_int() & 0xffff0000) != 0) && ((n->get_int() & 0xffff) == 0));
match(ConI);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
operand immInegpow2() %{
predicate(is_power_of_2_long((jlong) (julong) (juint) (-(n->get_int()))));
match(ConI);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
operand immIpow2minus1() %{
predicate(is_power_of_2_long((((jlong) (n->get_int()))+1)));
match(ConI);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
operand immIpowerOf2() %{
predicate(is_power_of_2_long((((jlong) (julong) (juint) (n->get_int())))));
match(ConI);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// Unsigned Integer Immediate: the values 0-31
operand uimmI5() %{
predicate(Assembler::is_uimm(n->get_int(), 5));
match(ConI);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// Unsigned Integer Immediate: 6-bit
operand uimmI6() %{
predicate(Assembler::is_uimm(n->get_int(), 6));
match(ConI);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// Unsigned Integer Immediate: 6-bit int, greater than 32
operand uimmI6_ge32() %{
predicate(Assembler::is_uimm(n->get_int(), 6) && n->get_int() >= 32);
match(ConI);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// Unsigned Integer Immediate: 15-bit
operand uimmI15() %{
predicate(Assembler::is_uimm(n->get_int(), 15));
match(ConI);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// Unsigned Integer Immediate: 16-bit
operand uimmI16() %{
predicate(Assembler::is_uimm(n->get_int(), 16));
match(ConI);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// constant 'int 0'.
operand immI_0() %{
predicate(n->get_int() == 0);
match(ConI);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// constant 'int 1'.
operand immI_1() %{
predicate(n->get_int() == 1);
match(ConI);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// constant 'int -1'.
operand immI_minus1() %{
predicate(n->get_int() == -1);
match(ConI);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// int value 16.
operand immI_16() %{
predicate(n->get_int() == 16);
match(ConI);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// int value 24.
operand immI_24() %{
predicate(n->get_int() == 24);
match(ConI);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// Compressed oops constants
// Pointer Immediate
operand immN() %{
match(ConN);
op_cost(10);
format %{ %}
interface(CONST_INTER);
%}
// NULL Pointer Immediate
operand immN_0() %{
predicate(n->get_narrowcon() == 0);
match(ConN);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// Compressed klass constants
operand immNKlass() %{
match(ConNKlass);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// This operand can be used to avoid matching of an instruct
// with chain rule.
operand immNKlass_NM() %{
match(ConNKlass);
predicate(false);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// Pointer Immediate: 64-bit
operand immP() %{
match(ConP);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// Operand to avoid match of loadConP.
// This operand can be used to avoid matching of an instruct
// with chain rule.
operand immP_NM() %{
match(ConP);
predicate(false);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// costant 'pointer 0'.
operand immP_0() %{
predicate(n->get_ptr() == 0);
match(ConP);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// pointer 0x0 or 0x1
operand immP_0or1() %{
predicate((n->get_ptr() == 0) || (n->get_ptr() == 1));
match(ConP);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
operand immL() %{
match(ConL);
op_cost(40);
format %{ %}
interface(CONST_INTER);
%}
// Long Immediate: 16-bit
operand immL16() %{
predicate(Assembler::is_simm(n->get_long(), 16));
match(ConL);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// Long Immediate: 16-bit, 4-aligned
operand immL16Alg4() %{
predicate(Assembler::is_simm(n->get_long(), 16) && ((n->get_long() & 0x3) == 0));
match(ConL);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// Long Immediate: 32-bit, where lowest 16 bits are 0x0000.
operand immL32hi16() %{
predicate(Assembler::is_simm(n->get_long(), 32) && ((n->get_long() & 0xffffL) == 0L));
match(ConL);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// Long Immediate: 32-bit
operand immL32() %{
predicate(Assembler::is_simm(n->get_long(), 32));
match(ConL);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// Long Immediate: 64-bit, where highest 16 bits are not 0x0000.
operand immLhighest16() %{
predicate((n->get_long() & 0xffff000000000000L) != 0L && (n->get_long() & 0x0000ffffffffffffL) == 0L);
match(ConL);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
operand immLnegpow2() %{
predicate(is_power_of_2_long((jlong)-(n->get_long())));
match(ConL);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
operand immLpow2minus1() %{
predicate(is_power_of_2_long((((jlong) (n->get_long()))+1)) &&
(n->get_long() != (jlong)0xffffffffffffffffL));
match(ConL);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// constant 'long 0'.
operand immL_0() %{
predicate(n->get_long() == 0L);
match(ConL);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// constat ' long -1'.
operand immL_minus1() %{
predicate(n->get_long() == -1L);
match(ConL);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// Long Immediate: low 32-bit mask
operand immL_32bits() %{
predicate(n->get_long() == 0xFFFFFFFFL);
match(ConL);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// Unsigned Long Immediate: 16-bit
operand uimmL16() %{
predicate(Assembler::is_uimm(n->get_long(), 16));
match(ConL);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// Float Immediate
operand immF() %{
match(ConF);
op_cost(40);
format %{ %}
interface(CONST_INTER);
%}
// Float Immediate: +0.0f.
operand immF_0() %{
predicate(jint_cast(n->getf()) == 0);
match(ConF);
op_cost(0);
format %{ %}
interface(CONST_INTER);
%}
// Double Immediate
operand immD() %{
match(ConD);
op_cost(40);
format %{ %}
interface(CONST_INTER);
%}
// Integer Register Operands
// Integer Destination Register
// See definition of reg_class bits32_reg_rw.
operand iRegIdst() %{
constraint(ALLOC_IN_RC(bits32_reg_rw));
match(RegI);
match(rscratch1RegI);
match(rscratch2RegI);
match(rarg1RegI);
match(rarg2RegI);
match(rarg3RegI);
match(rarg4RegI);
format %{ %}
interface(REG_INTER);
%}
// Integer Source Register
// See definition of reg_class bits32_reg_ro.
operand iRegIsrc() %{
constraint(ALLOC_IN_RC(bits32_reg_ro));
match(RegI);
match(rscratch1RegI);
match(rscratch2RegI);
match(rarg1RegI);
match(rarg2RegI);
match(rarg3RegI);
match(rarg4RegI);
format %{ %}
interface(REG_INTER);
%}
operand rscratch1RegI() %{
constraint(ALLOC_IN_RC(rscratch1_bits32_reg));
match(iRegIdst);
format %{ %}
interface(REG_INTER);
%}
operand rscratch2RegI() %{
constraint(ALLOC_IN_RC(rscratch2_bits32_reg));
match(iRegIdst);
format %{ %}
interface(REG_INTER);
%}
operand rarg1RegI() %{
constraint(ALLOC_IN_RC(rarg1_bits32_reg));
match(iRegIdst);
format %{ %}
interface(REG_INTER);
%}
operand rarg2RegI() %{
constraint(ALLOC_IN_RC(rarg2_bits32_reg));
match(iRegIdst);
format %{ %}
interface(REG_INTER);
%}
operand rarg3RegI() %{
constraint(ALLOC_IN_RC(rarg3_bits32_reg));
match(iRegIdst);
format %{ %}
interface(REG_INTER);
%}
operand rarg4RegI() %{
constraint(ALLOC_IN_RC(rarg4_bits32_reg));
match(iRegIdst);
format %{ %}
interface(REG_INTER);
%}
operand rarg1RegL() %{
constraint(ALLOC_IN_RC(rarg1_bits64_reg));
match(iRegLdst);
format %{ %}
interface(REG_INTER);
%}
operand rarg2RegL() %{
constraint(ALLOC_IN_RC(rarg2_bits64_reg));
match(iRegLdst);
format %{ %}
interface(REG_INTER);
%}
operand rarg3RegL() %{
constraint(ALLOC_IN_RC(rarg3_bits64_reg));
match(iRegLdst);
format %{ %}
interface(REG_INTER);
%}
operand rarg4RegL() %{
constraint(ALLOC_IN_RC(rarg4_bits64_reg));
match(iRegLdst);
format %{ %}
interface(REG_INTER);
%}
// Pointer Destination Register
// See definition of reg_class bits64_reg_rw.
operand iRegPdst() %{
constraint(ALLOC_IN_RC(bits64_reg_rw));
match(RegP);
match(rscratch1RegP);
match(rscratch2RegP);
match(rarg1RegP);
match(rarg2RegP);
match(rarg3RegP);
match(rarg4RegP);
format %{ %}
interface(REG_INTER);
%}
// Pointer Destination Register
// Operand not using r11 and r12 (killed in epilog).
operand iRegPdstNoScratch() %{
constraint(ALLOC_IN_RC(bits64_reg_leaf_call));
match(RegP);
match(rarg1RegP);
match(rarg2RegP);
match(rarg3RegP);
match(rarg4RegP);
format %{ %}
interface(REG_INTER);
%}
// Pointer Source Register
// See definition of reg_class bits64_reg_ro.
operand iRegPsrc() %{
constraint(ALLOC_IN_RC(bits64_reg_ro));
match(RegP);
match(iRegPdst);
match(rscratch1RegP);
match(rscratch2RegP);
match(rarg1RegP);
match(rarg2RegP);
match(rarg3RegP);
match(rarg4RegP);
match(threadRegP);
format %{ %}
interface(REG_INTER);
%}
// Thread operand.
operand threadRegP() %{
constraint(ALLOC_IN_RC(thread_bits64_reg));
match(iRegPdst);
format %{ "R16" %}
interface(REG_INTER);
%}
operand rscratch1RegP() %{
constraint(ALLOC_IN_RC(rscratch1_bits64_reg));
match(iRegPdst);
format %{ "R11" %}
interface(REG_INTER);
%}
operand rscratch2RegP() %{
constraint(ALLOC_IN_RC(rscratch2_bits64_reg));
match(iRegPdst);
format %{ %}
interface(REG_INTER);
%}
operand rarg1RegP() %{
constraint(ALLOC_IN_RC(rarg1_bits64_reg));
match(iRegPdst);
format %{ %}
interface(REG_INTER);
%}
operand rarg2RegP() %{
constraint(ALLOC_IN_RC(rarg2_bits64_reg));
match(iRegPdst);
format %{ %}
interface(REG_INTER);
%}
operand rarg3RegP() %{
constraint(ALLOC_IN_RC(rarg3_bits64_reg));
match(iRegPdst);
format %{ %}
interface(REG_INTER);
%}
operand rarg4RegP() %{
constraint(ALLOC_IN_RC(rarg4_bits64_reg));
match(iRegPdst);
format %{ %}
interface(REG_INTER);
%}
operand iRegNsrc() %{
constraint(ALLOC_IN_RC(bits32_reg_ro));
match(RegN);
match(iRegNdst);
format %{ %}
interface(REG_INTER);
%}
operand iRegNdst() %{
constraint(ALLOC_IN_RC(bits32_reg_rw));
match(RegN);
format %{ %}
interface(REG_INTER);
%}
// Long Destination Register
// See definition of reg_class bits64_reg_rw.
operand iRegLdst() %{
constraint(ALLOC_IN_RC(bits64_reg_rw));
match(RegL);
match(rscratch1RegL);
match(rscratch2RegL);
format %{ %}
interface(REG_INTER);
%}
// Long Source Register
// See definition of reg_class bits64_reg_ro.
operand iRegLsrc() %{
constraint(ALLOC_IN_RC(bits64_reg_ro));
match(RegL);
match(iRegLdst);
match(rscratch1RegL);
match(rscratch2RegL);
format %{ %}
interface(REG_INTER);
%}
// Special operand for ConvL2I.
operand iRegL2Isrc(iRegLsrc reg) %{
constraint(ALLOC_IN_RC(bits64_reg_ro));
match(ConvL2I reg);
format %{ "ConvL2I($reg)" %}
interface(REG_INTER)
%}
operand rscratch1RegL() %{
constraint(ALLOC_IN_RC(rscratch1_bits64_reg));
match(RegL);
format %{ %}
interface(REG_INTER);
%}
operand rscratch2RegL() %{
constraint(ALLOC_IN_RC(rscratch2_bits64_reg));
match(RegL);
format %{ %}
interface(REG_INTER);
%}
// Condition Code Flag Registers
operand flagsReg() %{
constraint(ALLOC_IN_RC(int_flags));
match(RegFlags);
format %{ %}
interface(REG_INTER);
%}
operand flagsRegSrc() %{
constraint(ALLOC_IN_RC(int_flags_ro));
match(RegFlags);
match(flagsReg);
match(flagsRegCR0);
format %{ %}
interface(REG_INTER);
%}
// Condition Code Flag Register CR0
operand flagsRegCR0() %{
constraint(ALLOC_IN_RC(int_flags_CR0));
match(RegFlags);
format %{ "CR0" %}
interface(REG_INTER);
%}
operand flagsRegCR1() %{
constraint(ALLOC_IN_RC(int_flags_CR1));
match(RegFlags);
format %{ "CR1" %}
interface(REG_INTER);
%}
operand flagsRegCR6() %{
constraint(ALLOC_IN_RC(int_flags_CR6));
match(RegFlags);
format %{ "CR6" %}
interface(REG_INTER);
%}
operand regCTR() %{
constraint(ALLOC_IN_RC(ctr_reg));
// RegFlags should work. Introducing a RegSpecial type would cause a
// lot of changes.
match(RegFlags);
format %{"SR_CTR" %}
interface(REG_INTER);
%}
operand regD() %{
constraint(ALLOC_IN_RC(dbl_reg));
match(RegD);
format %{ %}
interface(REG_INTER);
%}
operand regF() %{
constraint(ALLOC_IN_RC(flt_reg));
match(RegF);
format %{ %}
interface(REG_INTER);
%}
// Special Registers
// Method Register
operand inline_cache_regP(iRegPdst reg) %{
constraint(ALLOC_IN_RC(r19_bits64_reg)); // inline_cache_reg
match(reg);
format %{ %}
interface(REG_INTER);
%}
operand compiler_method_oop_regP(iRegPdst reg) %{
constraint(ALLOC_IN_RC(rscratch1_bits64_reg)); // compiler_method_oop_reg
match(reg);
format %{ %}
interface(REG_INTER);
%}
operand interpreter_method_oop_regP(iRegPdst reg) %{
constraint(ALLOC_IN_RC(r19_bits64_reg)); // interpreter_method_oop_reg
match(reg);
format %{ %}
interface(REG_INTER);
%}
// Operands to remove register moves in unscaled mode.
// Match read/write registers with an EncodeP node if neither shift nor add are required.
operand iRegP2N(iRegPsrc reg) %{
predicate(false /* TODO: PPC port MatchDecodeNodes*/&& Universe::narrow_oop_shift() == 0);
constraint(ALLOC_IN_RC(bits64_reg_ro));
match(EncodeP reg);
format %{ "$reg" %}
interface(REG_INTER)
%}
operand iRegN2P(iRegNsrc reg) %{
predicate(false /* TODO: PPC port MatchDecodeNodes*/);
constraint(ALLOC_IN_RC(bits32_reg_ro));
match(DecodeN reg);
format %{ "$reg" %}
interface(REG_INTER)
%}
operand iRegN2P_klass(iRegNsrc reg) %{
predicate(Universe::narrow_klass_base() == NULL && Universe::narrow_klass_shift() == 0);
constraint(ALLOC_IN_RC(bits32_reg_ro));
match(DecodeNKlass reg);
format %{ "$reg" %}
interface(REG_INTER)
%}
//----------Complex Operands---------------------------------------------------
// Indirect Memory Reference
operand indirect(iRegPsrc reg) %{
constraint(ALLOC_IN_RC(bits64_reg_ro));
match(reg);
op_cost(100);
format %{ "[$reg]" %}
interface(MEMORY_INTER) %{
base($reg);
index(0x0);
scale(0x0);
disp(0x0);
%}
%}
// Indirect with Offset
operand indOffset16(iRegPsrc reg, immL16 offset) %{
constraint(ALLOC_IN_RC(bits64_reg_ro));
match(AddP reg offset);
op_cost(100);
format %{ "[$reg + $offset]" %}
interface(MEMORY_INTER) %{
base($reg);
index(0x0);
scale(0x0);
disp($offset);
%}
%}
// Indirect with 4-aligned Offset
operand indOffset16Alg4(iRegPsrc reg, immL16Alg4 offset) %{
constraint(ALLOC_IN_RC(bits64_reg_ro));
match(AddP reg offset);
op_cost(100);
format %{ "[$reg + $offset]" %}
interface(MEMORY_INTER) %{
base($reg);
index(0x0);
scale(0x0);
disp($offset);
%}
%}
//----------Complex Operands for Compressed OOPs-------------------------------
// Compressed OOPs with narrow_oop_shift == 0.
// Indirect Memory Reference, compressed OOP
operand indirectNarrow(iRegNsrc reg) %{
predicate(false /* TODO: PPC port MatchDecodeNodes*/);
constraint(ALLOC_IN_RC(bits64_reg_ro));
match(DecodeN reg);
op_cost(100);
format %{ "[$reg]" %}
interface(MEMORY_INTER) %{
base($reg);
index(0x0);
scale(0x0);
disp(0x0);
%}
%}
operand indirectNarrow_klass(iRegNsrc reg) %{
predicate(Universe::narrow_klass_base() == NULL && Universe::narrow_klass_shift() == 0);
constraint(ALLOC_IN_RC(bits64_reg_ro));
match(DecodeNKlass reg);
op_cost(100);
format %{ "[$reg]" %}
interface(MEMORY_INTER) %{
base($reg);
index(0x0);
scale(0x0);
disp(0x0);
%}
%}
// Indirect with Offset, compressed OOP
operand indOffset16Narrow(iRegNsrc reg, immL16 offset) %{
predicate(false /* TODO: PPC port MatchDecodeNodes*/);
constraint(ALLOC_IN_RC(bits64_reg_ro));
match(AddP (DecodeN reg) offset);
op_cost(100);
format %{ "[$reg + $offset]" %}
interface(MEMORY_INTER) %{
base($reg);
index(0x0);
scale(0x0);
disp($offset);
%}
%}
operand indOffset16Narrow_klass(iRegNsrc reg, immL16 offset) %{
predicate(Universe::narrow_klass_base() == NULL && Universe::narrow_klass_shift() == 0);
constraint(ALLOC_IN_RC(bits64_reg_ro));
match(AddP (DecodeNKlass reg) offset);
op_cost(100);
format %{ "[$reg + $offset]" %}
interface(MEMORY_INTER) %{
base($reg);
index(0x0);
scale(0x0);
disp($offset);
%}
%}
// Indirect with 4-aligned Offset, compressed OOP
operand indOffset16NarrowAlg4(iRegNsrc reg, immL16Alg4 offset) %{
predicate(false /* TODO: PPC port MatchDecodeNodes*/);
constraint(ALLOC_IN_RC(bits64_reg_ro));
match(AddP (DecodeN reg) offset);
op_cost(100);
format %{ "[$reg + $offset]" %}
interface(MEMORY_INTER) %{
base($reg);
index(0x0);
scale(0x0);
disp($offset);
%}
%}
operand indOffset16NarrowAlg4_klass(iRegNsrc reg, immL16Alg4 offset) %{
predicate(Universe::narrow_klass_base() == NULL && Universe::narrow_klass_shift() == 0);
constraint(ALLOC_IN_RC(bits64_reg_ro));
match(AddP (DecodeNKlass reg) offset);
op_cost(100);
format %{ "[$reg + $offset]" %}
interface(MEMORY_INTER) %{
base($reg);
index(0x0);
scale(0x0);
disp($offset);
%}
%}
//----------Special Memory Operands--------------------------------------------
// Stack Slot Operand
//
// This operand is used for loading and storing temporary values on
// the stack where a match requires a value to flow through memory.
operand stackSlotI(sRegI reg) %{
constraint(ALLOC_IN_RC(stack_slots));
op_cost(100);
//match(RegI);
format %{ "[sp+$reg]" %}
interface(MEMORY_INTER) %{
base(0x1); // R1_SP
index(0x0);
scale(0x0);
disp($reg); // Stack Offset
%}
%}
operand stackSlotL(sRegL reg) %{
constraint(ALLOC_IN_RC(stack_slots));
op_cost(100);
//match(RegL);
format %{ "[sp+$reg]" %}
interface(MEMORY_INTER) %{
base(0x1); // R1_SP
index(0x0);
scale(0x0);
disp($reg); // Stack Offset
%}
%}
operand stackSlotP(sRegP reg) %{
constraint(ALLOC_IN_RC(stack_slots));
op_cost(100);
//match(RegP);
format %{ "[sp+$reg]" %}
interface(MEMORY_INTER) %{
base(0x1); // R1_SP
index(0x0);
scale(0x0);
disp($reg); // Stack Offset
%}
%}
operand stackSlotF(sRegF reg) %{
constraint(ALLOC_IN_RC(stack_slots));
op_cost(100);
//match(RegF);
format %{ "[sp+$reg]" %}
interface(MEMORY_INTER) %{
base(0x1); // R1_SP
index(0x0);
scale(0x0);
disp($reg); // Stack Offset
%}
%}
operand stackSlotD(sRegD reg) %{
constraint(ALLOC_IN_RC(stack_slots));
op_cost(100);
//match(RegD);
format %{ "[sp+$reg]" %}
interface(MEMORY_INTER) %{
base(0x1); // R1_SP
index(0x0);
scale(0x0);
disp($reg); // Stack Offset
%}
%}
// Operands for expressing Control Flow
// NOTE: Label is a predefined operand which should not be redefined in
// the AD file. It is generically handled within the ADLC.
//----------Conditional Branch Operands----------------------------------------
// Comparison Op
//
// This is the operation of the comparison, and is limited to the
// following set of codes: L (<), LE (<=), G (>), GE (>=), E (==), NE
// (!=).
//
// Other attributes of the comparison, such as unsignedness, are specified
// by the comparison instruction that sets a condition code flags register.
// That result is represented by a flags operand whose subtype is appropriate
// to the unsignedness (etc.) of the comparison.
//
// Later, the instruction which matches both the Comparison Op (a Bool) and
// the flags (produced by the Cmp) specifies the coding of the comparison op
// by matching a specific subtype of Bool operand below.
// When used for floating point comparisons: unordered same as less.
operand cmpOp() %{
match(Bool);
format %{ "" %}
interface(COND_INTER) %{
// BO only encodes bit 4 of bcondCRbiIsX, as bits 1-3 are always '100'.
// BO & BI
equal(0xA); // 10 10: bcondCRbiIs1 & Condition::equal
not_equal(0x2); // 00 10: bcondCRbiIs0 & Condition::equal
less(0x8); // 10 00: bcondCRbiIs1 & Condition::less
greater_equal(0x0); // 00 00: bcondCRbiIs0 & Condition::less
less_equal(0x1); // 00 01: bcondCRbiIs0 & Condition::greater
greater(0x9); // 10 01: bcondCRbiIs1 & Condition::greater
overflow(0xB); // 10 11: bcondCRbiIs1 & Condition::summary_overflow
no_overflow(0x3); // 00 11: bcondCRbiIs0 & Condition::summary_overflow
%}
%}
//----------OPERAND CLASSES----------------------------------------------------
// Operand Classes are groups of operands that are used to simplify
// instruction definitions by not requiring the AD writer to specify
// seperate instructions for every form of operand when the
// instruction accepts multiple operand types with the same basic
// encoding and format. The classic case of this is memory operands.
// Indirect is not included since its use is limited to Compare & Swap.
opclass memory(indirect, indOffset16 /*, indIndex, tlsReference*/, indirectNarrow, indirectNarrow_klass, indOffset16Narrow, indOffset16Narrow_klass);
// Memory operand where offsets are 4-aligned. Required for ld, std.
opclass memoryAlg4(indirect, indOffset16Alg4, indirectNarrow, indOffset16NarrowAlg4, indOffset16NarrowAlg4_klass);
opclass indirectMemory(indirect, indirectNarrow);
// Special opclass for I and ConvL2I.
opclass iRegIsrc_iRegL2Isrc(iRegIsrc, iRegL2Isrc);
// Operand classes to match encode and decode. iRegN_P2N is only used
// for storeN. I have never seen an encode node elsewhere.
opclass iRegN_P2N(iRegNsrc, iRegP2N);
opclass iRegP_N2P(iRegPsrc, iRegN2P, iRegN2P_klass);
//----------PIPELINE-----------------------------------------------------------
pipeline %{
// See J.M.Tendler et al. "Power4 system microarchitecture", IBM
// J. Res. & Dev., No. 1, Jan. 2002.
//----------ATTRIBUTES---------------------------------------------------------
attributes %{
// Power4 instructions are of fixed length.
fixed_size_instructions;
// TODO: if `bundle' means number of instructions fetched
// per cycle, this is 8. If `bundle' means Power4 `group', that is
// max instructions issued per cycle, this is 5.
max_instructions_per_bundle = 8;
// A Power4 instruction is 4 bytes long.
instruction_unit_size = 4;
// The Power4 processor fetches 64 bytes...
instruction_fetch_unit_size = 64;
// ...in one line
instruction_fetch_units = 1
// Unused, list one so that array generated by adlc is not empty.
// Aix compiler chokes if _nop_count = 0.
nops(fxNop);
%}
//----------RESOURCES----------------------------------------------------------
// Resources are the functional units available to the machine
resources(
PPC_BR, // branch unit
PPC_CR, // condition unit
PPC_FX1, // integer arithmetic unit 1
PPC_FX2, // integer arithmetic unit 2
PPC_LDST1, // load/store unit 1
PPC_LDST2, // load/store unit 2
PPC_FP1, // float arithmetic unit 1
PPC_FP2, // float arithmetic unit 2
PPC_LDST = PPC_LDST1 | PPC_LDST2,
PPC_FX = PPC_FX1 | PPC_FX2,
PPC_FP = PPC_FP1 | PPC_FP2
);
//----------PIPELINE DESCRIPTION-----------------------------------------------
// Pipeline Description specifies the stages in the machine's pipeline
pipe_desc(
// Power4 longest pipeline path
PPC_IF, // instruction fetch
PPC_IC,
//PPC_BP, // branch prediction
PPC_D0, // decode
PPC_D1, // decode
PPC_D2, // decode
PPC_D3, // decode
PPC_Xfer1,
PPC_GD, // group definition
PPC_MP, // map
PPC_ISS, // issue
PPC_RF, // resource fetch
PPC_EX1, // execute (all units)
PPC_EX2, // execute (FP, LDST)
PPC_EX3, // execute (FP, LDST)
PPC_EX4, // execute (FP)
PPC_EX5, // execute (FP)
PPC_EX6, // execute (FP)
PPC_WB, // write back
PPC_Xfer2,
PPC_CP
);
//----------PIPELINE CLASSES---------------------------------------------------
// Pipeline Classes describe the stages in which input and output are
// referenced by the hardware pipeline.
// Simple pipeline classes.
// Default pipeline class.
pipe_class pipe_class_default() %{
single_instruction;
fixed_latency(2);
%}
// Pipeline class for empty instructions.
pipe_class pipe_class_empty() %{
single_instruction;
fixed_latency(0);
%}
// Pipeline class for compares.
pipe_class pipe_class_compare() %{
single_instruction;
fixed_latency(16);
%}
// Pipeline class for traps.
pipe_class pipe_class_trap() %{
single_instruction;
fixed_latency(100);
%}
// Pipeline class for memory operations.
pipe_class pipe_class_memory() %{
single_instruction;
fixed_latency(16);
%}
// Pipeline class for call.
pipe_class pipe_class_call() %{
single_instruction;
fixed_latency(100);
%}
// Define the class for the Nop node.
define %{
MachNop = pipe_class_default;
%}
%}
//----------INSTRUCTIONS-------------------------------------------------------
// Naming of instructions:
// opA_operB / opA_operB_operC:
// Operation 'op' with one or two source operands 'oper'. Result
// type is A, source operand types are B and C.
// Iff A == B == C, B and C are left out.
//
// The instructions are ordered according to the following scheme:
// - loads
// - load constants
// - prefetch
// - store
// - encode/decode
// - membar
// - conditional moves
// - compare & swap
// - arithmetic and logic operations
// * int: Add, Sub, Mul, Div, Mod
// * int: lShift, arShift, urShift, rot
// * float: Add, Sub, Mul, Div
// * and, or, xor ...
// - register moves: float <-> int, reg <-> stack, repl
// - cast (high level type cast, XtoP, castPP, castII, not_null etc.
// - conv (low level type cast requiring bit changes (sign extend etc)
// - compares, range & zero checks.
// - branches
// - complex operations, intrinsics, min, max, replicate
// - lock
// - Calls
//
// If there are similar instructions with different types they are sorted:
// int before float
// small before big
// signed before unsigned
// e.g., loadS before loadUS before loadI before loadF.
//----------Load/Store Instructions--------------------------------------------
//----------Load Instructions--------------------------------------------------
// Converts byte to int.
// As convB2I_reg, but without match rule. The match rule of convB2I_reg
// reuses the 'amount' operand, but adlc expects that operand specification
// and operands in match rule are equivalent.
instruct convB2I_reg_2(iRegIdst dst, iRegIsrc src) %{
effect(DEF dst, USE src);
format %{ "EXTSB $dst, $src \t// byte->int" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_extsb);
__ extsb($dst$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct loadUB_indirect(iRegIdst dst, indirectMemory mem) %{
// match-rule, false predicate
match(Set dst (LoadB mem));
predicate(false);
format %{ "LBZ $dst, $mem" %}
size(4);
ins_encode( enc_lbz(dst, mem) );
ins_pipe(pipe_class_memory);
%}
instruct loadUB_indirect_ac(iRegIdst dst, indirectMemory mem) %{
// match-rule, false predicate
match(Set dst (LoadB mem));
predicate(false);
format %{ "LBZ $dst, $mem\n\t"
"TWI $dst\n\t"
"ISYNC" %}
size(12);
ins_encode( enc_lbz_ac(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Byte (8bit signed). LoadB = LoadUB + ConvUB2B.
instruct loadB_indirect_Ex(iRegIdst dst, indirectMemory mem) %{
match(Set dst (LoadB mem));
predicate(n->as_Load()->is_unordered() || followed_by_acquire(n));
ins_cost(MEMORY_REF_COST + DEFAULT_COST);
expand %{
iRegIdst tmp;
loadUB_indirect(tmp, mem);
convB2I_reg_2(dst, tmp);
%}
%}
instruct loadB_indirect_ac_Ex(iRegIdst dst, indirectMemory mem) %{
match(Set dst (LoadB mem));
ins_cost(3*MEMORY_REF_COST + DEFAULT_COST);
expand %{
iRegIdst tmp;
loadUB_indirect_ac(tmp, mem);
convB2I_reg_2(dst, tmp);
%}
%}
instruct loadUB_indOffset16(iRegIdst dst, indOffset16 mem) %{
// match-rule, false predicate
match(Set dst (LoadB mem));
predicate(false);
format %{ "LBZ $dst, $mem" %}
size(4);
ins_encode( enc_lbz(dst, mem) );
ins_pipe(pipe_class_memory);
%}
instruct loadUB_indOffset16_ac(iRegIdst dst, indOffset16 mem) %{
// match-rule, false predicate
match(Set dst (LoadB mem));
predicate(false);
format %{ "LBZ $dst, $mem\n\t"
"TWI $dst\n\t"
"ISYNC" %}
size(12);
ins_encode( enc_lbz_ac(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Byte (8bit signed). LoadB = LoadUB + ConvUB2B.
instruct loadB_indOffset16_Ex(iRegIdst dst, indOffset16 mem) %{
match(Set dst (LoadB mem));
predicate(n->as_Load()->is_unordered() || followed_by_acquire(n));
ins_cost(MEMORY_REF_COST + DEFAULT_COST);
expand %{
iRegIdst tmp;
loadUB_indOffset16(tmp, mem);
convB2I_reg_2(dst, tmp);
%}
%}
instruct loadB_indOffset16_ac_Ex(iRegIdst dst, indOffset16 mem) %{
match(Set dst (LoadB mem));
ins_cost(3*MEMORY_REF_COST + DEFAULT_COST);
expand %{
iRegIdst tmp;
loadUB_indOffset16_ac(tmp, mem);
convB2I_reg_2(dst, tmp);
%}
%}
// Load Unsigned Byte (8bit UNsigned) into an int reg.
instruct loadUB(iRegIdst dst, memory mem) %{
predicate(n->as_Load()->is_unordered() || followed_by_acquire(n));
match(Set dst (LoadUB mem));
ins_cost(MEMORY_REF_COST);
format %{ "LBZ $dst, $mem \t// byte, zero-extend to int" %}
size(4);
ins_encode( enc_lbz(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Unsigned Byte (8bit UNsigned) acquire.
instruct loadUB_ac(iRegIdst dst, memory mem) %{
match(Set dst (LoadUB mem));
ins_cost(3*MEMORY_REF_COST);
format %{ "LBZ $dst, $mem \t// byte, zero-extend to int, acquire\n\t"
"TWI $dst\n\t"
"ISYNC" %}
size(12);
ins_encode( enc_lbz_ac(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Unsigned Byte (8bit UNsigned) into a Long Register.
instruct loadUB2L(iRegLdst dst, memory mem) %{
match(Set dst (ConvI2L (LoadUB mem)));
predicate(_kids[0]->_leaf->as_Load()->is_unordered() || followed_by_acquire(_kids[0]->_leaf));
ins_cost(MEMORY_REF_COST);
format %{ "LBZ $dst, $mem \t// byte, zero-extend to long" %}
size(4);
ins_encode( enc_lbz(dst, mem) );
ins_pipe(pipe_class_memory);
%}
instruct loadUB2L_ac(iRegLdst dst, memory mem) %{
match(Set dst (ConvI2L (LoadUB mem)));
ins_cost(3*MEMORY_REF_COST);
format %{ "LBZ $dst, $mem \t// byte, zero-extend to long, acquire\n\t"
"TWI $dst\n\t"
"ISYNC" %}
size(12);
ins_encode( enc_lbz_ac(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Short (16bit signed)
instruct loadS(iRegIdst dst, memory mem) %{
match(Set dst (LoadS mem));
predicate(n->as_Load()->is_unordered() || followed_by_acquire(n));
ins_cost(MEMORY_REF_COST);
format %{ "LHA $dst, $mem" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_lha);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
__ lha($dst$$Register, Idisp, $mem$$base$$Register);
%}
ins_pipe(pipe_class_memory);
%}
// Load Short (16bit signed) acquire.
instruct loadS_ac(iRegIdst dst, memory mem) %{
match(Set dst (LoadS mem));
ins_cost(3*MEMORY_REF_COST);
format %{ "LHA $dst, $mem\t acquire\n\t"
"TWI $dst\n\t"
"ISYNC" %}
size(12);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
__ lha($dst$$Register, Idisp, $mem$$base$$Register);
__ twi_0($dst$$Register);
__ isync();
%}
ins_pipe(pipe_class_memory);
%}
// Load Char (16bit unsigned)
instruct loadUS(iRegIdst dst, memory mem) %{
match(Set dst (LoadUS mem));
predicate(n->as_Load()->is_unordered() || followed_by_acquire(n));
ins_cost(MEMORY_REF_COST);
format %{ "LHZ $dst, $mem" %}
size(4);
ins_encode( enc_lhz(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Char (16bit unsigned) acquire.
instruct loadUS_ac(iRegIdst dst, memory mem) %{
match(Set dst (LoadUS mem));
ins_cost(3*MEMORY_REF_COST);
format %{ "LHZ $dst, $mem \t// acquire\n\t"
"TWI $dst\n\t"
"ISYNC" %}
size(12);
ins_encode( enc_lhz_ac(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Unsigned Short/Char (16bit UNsigned) into a Long Register.
instruct loadUS2L(iRegLdst dst, memory mem) %{
match(Set dst (ConvI2L (LoadUS mem)));
predicate(_kids[0]->_leaf->as_Load()->is_unordered() || followed_by_acquire(_kids[0]->_leaf));
ins_cost(MEMORY_REF_COST);
format %{ "LHZ $dst, $mem \t// short, zero-extend to long" %}
size(4);
ins_encode( enc_lhz(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Unsigned Short/Char (16bit UNsigned) into a Long Register acquire.
instruct loadUS2L_ac(iRegLdst dst, memory mem) %{
match(Set dst (ConvI2L (LoadUS mem)));
ins_cost(3*MEMORY_REF_COST);
format %{ "LHZ $dst, $mem \t// short, zero-extend to long, acquire\n\t"
"TWI $dst\n\t"
"ISYNC" %}
size(12);
ins_encode( enc_lhz_ac(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Integer.
instruct loadI(iRegIdst dst, memory mem) %{
match(Set dst (LoadI mem));
predicate(n->as_Load()->is_unordered() || followed_by_acquire(n));
ins_cost(MEMORY_REF_COST);
format %{ "LWZ $dst, $mem" %}
size(4);
ins_encode( enc_lwz(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Integer acquire.
instruct loadI_ac(iRegIdst dst, memory mem) %{
match(Set dst (LoadI mem));
ins_cost(3*MEMORY_REF_COST);
format %{ "LWZ $dst, $mem \t// load acquire\n\t"
"TWI $dst\n\t"
"ISYNC" %}
size(12);
ins_encode( enc_lwz_ac(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Match loading integer and casting it to unsigned int in
// long register.
// LoadI + ConvI2L + AndL 0xffffffff.
instruct loadUI2L(iRegLdst dst, memory mem, immL_32bits mask) %{
match(Set dst (AndL (ConvI2L (LoadI mem)) mask));
predicate(_kids[0]->_kids[0]->_leaf->as_Load()->is_unordered());
ins_cost(MEMORY_REF_COST);
format %{ "LWZ $dst, $mem \t// zero-extend to long" %}
size(4);
ins_encode( enc_lwz(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Match loading integer and casting it to long.
instruct loadI2L(iRegLdst dst, memory mem) %{
match(Set dst (ConvI2L (LoadI mem)));
predicate(_kids[0]->_leaf->as_Load()->is_unordered());
ins_cost(MEMORY_REF_COST);
format %{ "LWA $dst, $mem \t// loadI2L" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_lwa);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
__ lwa($dst$$Register, Idisp, $mem$$base$$Register);
%}
ins_pipe(pipe_class_memory);
%}
// Match loading integer and casting it to long - acquire.
instruct loadI2L_ac(iRegLdst dst, memory mem) %{
match(Set dst (ConvI2L (LoadI mem)));
ins_cost(3*MEMORY_REF_COST);
format %{ "LWA $dst, $mem \t// loadI2L acquire"
"TWI $dst\n\t"
"ISYNC" %}
size(12);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_lwa);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
__ lwa($dst$$Register, Idisp, $mem$$base$$Register);
__ twi_0($dst$$Register);
__ isync();
%}
ins_pipe(pipe_class_memory);
%}
// Load Long - aligned
instruct loadL(iRegLdst dst, memoryAlg4 mem) %{
match(Set dst (LoadL mem));
predicate(n->as_Load()->is_unordered() || followed_by_acquire(n));
ins_cost(MEMORY_REF_COST);
format %{ "LD $dst, $mem \t// long" %}
size(4);
ins_encode( enc_ld(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Long - aligned acquire.
instruct loadL_ac(iRegLdst dst, memoryAlg4 mem) %{
match(Set dst (LoadL mem));
ins_cost(3*MEMORY_REF_COST);
format %{ "LD $dst, $mem \t// long acquire\n\t"
"TWI $dst\n\t"
"ISYNC" %}
size(12);
ins_encode( enc_ld_ac(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Long - UNaligned
instruct loadL_unaligned(iRegLdst dst, memoryAlg4 mem) %{
match(Set dst (LoadL_unaligned mem));
// predicate(...) // Unaligned_ac is not needed (and wouldn't make sense).
ins_cost(MEMORY_REF_COST);
format %{ "LD $dst, $mem \t// unaligned long" %}
size(4);
ins_encode( enc_ld(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load nodes for superwords
// Load Aligned Packed Byte
instruct loadV8(iRegLdst dst, memoryAlg4 mem) %{
predicate(n->as_LoadVector()->memory_size() == 8);
match(Set dst (LoadVector mem));
ins_cost(MEMORY_REF_COST);
format %{ "LD $dst, $mem \t// load 8-byte Vector" %}
size(4);
ins_encode( enc_ld(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Range, range = array length (=jint)
instruct loadRange(iRegIdst dst, memory mem) %{
match(Set dst (LoadRange mem));
ins_cost(MEMORY_REF_COST);
format %{ "LWZ $dst, $mem \t// range" %}
size(4);
ins_encode( enc_lwz(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Compressed Pointer
instruct loadN(iRegNdst dst, memory mem) %{
match(Set dst (LoadN mem));
predicate(n->as_Load()->is_unordered() || followed_by_acquire(n));
ins_cost(MEMORY_REF_COST);
format %{ "LWZ $dst, $mem \t// load compressed ptr" %}
size(4);
ins_encode( enc_lwz(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Compressed Pointer acquire.
instruct loadN_ac(iRegNdst dst, memory mem) %{
match(Set dst (LoadN mem));
ins_cost(3*MEMORY_REF_COST);
format %{ "LWZ $dst, $mem \t// load acquire compressed ptr\n\t"
"TWI $dst\n\t"
"ISYNC" %}
size(12);
ins_encode( enc_lwz_ac(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Compressed Pointer and decode it if narrow_oop_shift == 0.
instruct loadN2P_unscaled(iRegPdst dst, memory mem) %{
match(Set dst (DecodeN (LoadN mem)));
predicate(_kids[0]->_leaf->as_Load()->is_unordered() && Universe::narrow_oop_shift() == 0);
ins_cost(MEMORY_REF_COST);
format %{ "LWZ $dst, $mem \t// DecodeN (unscaled)" %}
size(4);
ins_encode( enc_lwz(dst, mem) );
ins_pipe(pipe_class_memory);
%}
instruct loadN2P_klass_unscaled(iRegPdst dst, memory mem) %{
match(Set dst (DecodeNKlass (LoadNKlass mem)));
predicate(Universe::narrow_klass_base() == NULL && Universe::narrow_klass_shift() == 0 &&
_kids[0]->_leaf->as_Load()->is_unordered());
ins_cost(MEMORY_REF_COST);
format %{ "LWZ $dst, $mem \t// DecodeN (unscaled)" %}
size(4);
ins_encode( enc_lwz(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Pointer
instruct loadP(iRegPdst dst, memoryAlg4 mem) %{
match(Set dst (LoadP mem));
predicate(n->as_Load()->is_unordered() || followed_by_acquire(n));
ins_cost(MEMORY_REF_COST);
format %{ "LD $dst, $mem \t// ptr" %}
size(4);
ins_encode( enc_ld(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Pointer acquire.
instruct loadP_ac(iRegPdst dst, memoryAlg4 mem) %{
match(Set dst (LoadP mem));
ins_cost(3*MEMORY_REF_COST);
format %{ "LD $dst, $mem \t// ptr acquire\n\t"
"TWI $dst\n\t"
"ISYNC" %}
size(12);
ins_encode( enc_ld_ac(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// LoadP + CastP2L
instruct loadP2X(iRegLdst dst, memoryAlg4 mem) %{
match(Set dst (CastP2X (LoadP mem)));
predicate(_kids[0]->_leaf->as_Load()->is_unordered());
ins_cost(MEMORY_REF_COST);
format %{ "LD $dst, $mem \t// ptr + p2x" %}
size(4);
ins_encode( enc_ld(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load compressed klass pointer.
instruct loadNKlass(iRegNdst dst, memory mem) %{
match(Set dst (LoadNKlass mem));
ins_cost(MEMORY_REF_COST);
format %{ "LWZ $dst, $mem \t// compressed klass ptr" %}
size(4);
ins_encode( enc_lwz(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Klass Pointer
instruct loadKlass(iRegPdst dst, memoryAlg4 mem) %{
match(Set dst (LoadKlass mem));
ins_cost(MEMORY_REF_COST);
format %{ "LD $dst, $mem \t// klass ptr" %}
size(4);
ins_encode( enc_ld(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Float
instruct loadF(regF dst, memory mem) %{
match(Set dst (LoadF mem));
predicate(n->as_Load()->is_unordered() || followed_by_acquire(n));
ins_cost(MEMORY_REF_COST);
format %{ "LFS $dst, $mem" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_lfs);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
__ lfs($dst$$FloatRegister, Idisp, $mem$$base$$Register);
%}
ins_pipe(pipe_class_memory);
%}
// Load Float acquire.
instruct loadF_ac(regF dst, memory mem, flagsRegCR0 cr0) %{
match(Set dst (LoadF mem));
effect(TEMP cr0);
ins_cost(3*MEMORY_REF_COST);
format %{ "LFS $dst, $mem \t// acquire\n\t"
"FCMPU cr0, $dst, $dst\n\t"
"BNE cr0, next\n"
"next:\n\t"
"ISYNC" %}
size(16);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
Label next;
__ lfs($dst$$FloatRegister, Idisp, $mem$$base$$Register);
__ fcmpu(CCR0, $dst$$FloatRegister, $dst$$FloatRegister);
__ bne(CCR0, next);
__ bind(next);
__ isync();
%}
ins_pipe(pipe_class_memory);
%}
// Load Double - aligned
instruct loadD(regD dst, memory mem) %{
match(Set dst (LoadD mem));
predicate(n->as_Load()->is_unordered() || followed_by_acquire(n));
ins_cost(MEMORY_REF_COST);
format %{ "LFD $dst, $mem" %}
size(4);
ins_encode( enc_lfd(dst, mem) );
ins_pipe(pipe_class_memory);
%}
// Load Double - aligned acquire.
instruct loadD_ac(regD dst, memory mem, flagsRegCR0 cr0) %{
match(Set dst (LoadD mem));
effect(TEMP cr0);
ins_cost(3*MEMORY_REF_COST);
format %{ "LFD $dst, $mem \t// acquire\n\t"
"FCMPU cr0, $dst, $dst\n\t"
"BNE cr0, next\n"
"next:\n\t"
"ISYNC" %}
size(16);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
Label next;
__ lfd($dst$$FloatRegister, Idisp, $mem$$base$$Register);
__ fcmpu(CCR0, $dst$$FloatRegister, $dst$$FloatRegister);
__ bne(CCR0, next);
__ bind(next);
__ isync();
%}
ins_pipe(pipe_class_memory);
%}
// Load Double - UNaligned
instruct loadD_unaligned(regD dst, memory mem) %{
match(Set dst (LoadD_unaligned mem));
// predicate(...) // Unaligned_ac is not needed (and wouldn't make sense).
ins_cost(MEMORY_REF_COST);
format %{ "LFD $dst, $mem" %}
size(4);
ins_encode( enc_lfd(dst, mem) );
ins_pipe(pipe_class_memory);
%}
//----------Constants--------------------------------------------------------
// Load MachConstantTableBase: add hi offset to global toc.
// TODO: Handle hidden register r29 in bundler!
instruct loadToc_hi(iRegLdst dst) %{
effect(DEF dst);
ins_cost(DEFAULT_COST);
format %{ "ADDIS $dst, R29, DISP.hi \t// load TOC hi" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addis);
__ calculate_address_from_global_toc_hi16only($dst$$Register, __ method_toc());
%}
ins_pipe(pipe_class_default);
%}
// Load MachConstantTableBase: add lo offset to global toc.
instruct loadToc_lo(iRegLdst dst, iRegLdst src) %{
effect(DEF dst, USE src);
ins_cost(DEFAULT_COST);
format %{ "ADDI $dst, $src, DISP.lo \t// load TOC lo" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_ori);
__ calculate_address_from_global_toc_lo16only($dst$$Register, __ method_toc());
%}
ins_pipe(pipe_class_default);
%}
// Load 16-bit integer constant 0xssss????
instruct loadConI16(iRegIdst dst, immI16 src) %{
match(Set dst src);
format %{ "LI $dst, $src" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addi);
__ li($dst$$Register, (int)((short)($src$$constant & 0xFFFF)));
%}
ins_pipe(pipe_class_default);
%}
// Load integer constant 0x????0000
instruct loadConIhi16(iRegIdst dst, immIhi16 src) %{
match(Set dst src);
ins_cost(DEFAULT_COST);
format %{ "LIS $dst, $src.hi" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addis);
// Lis sign extends 16-bit src then shifts it 16 bit to the left.
__ lis($dst$$Register, (int)((short)(($src$$constant & 0xFFFF0000) >> 16)));
%}
ins_pipe(pipe_class_default);
%}
// Part 2 of loading 32 bit constant: hi16 is is src1 (properly shifted
// and sign extended), this adds the low 16 bits.
instruct loadConI32_lo16(iRegIdst dst, iRegIsrc src1, immI16 src2) %{
// no match-rule, false predicate
effect(DEF dst, USE src1, USE src2);
predicate(false);
format %{ "ORI $dst, $src1.hi, $src2.lo" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_ori);
__ ori($dst$$Register, $src1$$Register, ($src2$$constant) & 0xFFFF);
%}
ins_pipe(pipe_class_default);
%}
instruct loadConI_Ex(iRegIdst dst, immI src) %{
match(Set dst src);
ins_cost(DEFAULT_COST*2);
expand %{
// Would like to use $src$$constant.
immI16 srcLo %{ _opnds[1]->constant() %}
// srcHi can be 0000 if srcLo sign-extends to a negative number.
immIhi16 srcHi %{ _opnds[1]->constant() %}
iRegIdst tmpI;
loadConIhi16(tmpI, srcHi);
loadConI32_lo16(dst, tmpI, srcLo);
%}
%}
// No constant pool entries required.
instruct loadConL16(iRegLdst dst, immL16 src) %{
match(Set dst src);
format %{ "LI $dst, $src \t// long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addi);
__ li($dst$$Register, (int)((short) ($src$$constant & 0xFFFF)));
%}
ins_pipe(pipe_class_default);
%}
// Load long constant 0xssssssss????0000
instruct loadConL32hi16(iRegLdst dst, immL32hi16 src) %{
match(Set dst src);
ins_cost(DEFAULT_COST);
format %{ "LIS $dst, $src.hi \t// long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addis);
__ lis($dst$$Register, (int)((short)(($src$$constant & 0xFFFF0000) >> 16)));
%}
ins_pipe(pipe_class_default);
%}
// To load a 32 bit constant: merge lower 16 bits into already loaded
// high 16 bits.
instruct loadConL32_lo16(iRegLdst dst, iRegLsrc src1, immL16 src2) %{
// no match-rule, false predicate
effect(DEF dst, USE src1, USE src2);
predicate(false);
format %{ "ORI $dst, $src1, $src2.lo" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_ori);
__ ori($dst$$Register, $src1$$Register, ($src2$$constant) & 0xFFFF);
%}
ins_pipe(pipe_class_default);
%}
// Load 32-bit long constant
instruct loadConL32_Ex(iRegLdst dst, immL32 src) %{
match(Set dst src);
ins_cost(DEFAULT_COST*2);
expand %{
// Would like to use $src$$constant.
immL16 srcLo %{ _opnds[1]->constant() /*& 0x0000FFFFL */%}
// srcHi can be 0000 if srcLo sign-extends to a negative number.
immL32hi16 srcHi %{ _opnds[1]->constant() /*& 0xFFFF0000L */%}
iRegLdst tmpL;
loadConL32hi16(tmpL, srcHi);
loadConL32_lo16(dst, tmpL, srcLo);
%}
%}
// Load long constant 0x????000000000000.
instruct loadConLhighest16_Ex(iRegLdst dst, immLhighest16 src) %{
match(Set dst src);
ins_cost(DEFAULT_COST);
expand %{
immL32hi16 srcHi %{ _opnds[1]->constant() >> 32 /*& 0xFFFF0000L */%}
immI shift32 %{ 32 %}
iRegLdst tmpL;
loadConL32hi16(tmpL, srcHi);
lshiftL_regL_immI(dst, tmpL, shift32);
%}
%}
// Expand node for constant pool load: small offset.
instruct loadConL(iRegLdst dst, immL src, iRegLdst toc) %{
effect(DEF dst, USE src, USE toc);
ins_cost(MEMORY_REF_COST);
ins_num_consts(1);
// Needed so that CallDynamicJavaDirect can compute the address of this
// instruction for relocation.
ins_field_cbuf_insts_offset(int);
format %{ "LD $dst, offset, $toc \t// load long $src from TOC" %}
size(4);
ins_encode( enc_load_long_constL(dst, src, toc) );
ins_pipe(pipe_class_memory);
%}
// Expand node for constant pool load: large offset.
instruct loadConL_hi(iRegLdst dst, immL src, iRegLdst toc) %{
effect(DEF dst, USE src, USE toc);
predicate(false);
ins_num_consts(1);
ins_field_const_toc_offset(int);
// Needed so that CallDynamicJavaDirect can compute the address of this
// instruction for relocation.
ins_field_cbuf_insts_offset(int);
format %{ "ADDIS $dst, $toc, offset \t// load long $src from TOC (hi)" %}
size(4);
ins_encode( enc_load_long_constL_hi(dst, toc, src) );
ins_pipe(pipe_class_default);
%}
// Expand node for constant pool load: large offset.
// No constant pool entries required.
instruct loadConL_lo(iRegLdst dst, immL src, iRegLdst base) %{
effect(DEF dst, USE src, USE base);
predicate(false);
ins_field_const_toc_offset_hi_node(loadConL_hiNode*);
format %{ "LD $dst, offset, $base \t// load long $src from TOC (lo)" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_ld);
int offset = ra_->C->in_scratch_emit_size() ? 0 : _const_toc_offset_hi_node->_const_toc_offset;
__ ld($dst$$Register, MacroAssembler::largeoffset_si16_si16_lo(offset), $base$$Register);
%}
ins_pipe(pipe_class_memory);
%}
// Load long constant from constant table. Expand in case of
// offset > 16 bit is needed.
// Adlc adds toc node MachConstantTableBase.
instruct loadConL_Ex(iRegLdst dst, immL src) %{
match(Set dst src);
ins_cost(MEMORY_REF_COST);
format %{ "LD $dst, offset, $constanttablebase\t// load long $src from table, postalloc expanded" %}
// We can not inline the enc_class for the expand as that does not support constanttablebase.
postalloc_expand( postalloc_expand_load_long_constant(dst, src, constanttablebase) );
%}
// Load NULL as compressed oop.
instruct loadConN0(iRegNdst dst, immN_0 src) %{
match(Set dst src);
ins_cost(DEFAULT_COST);
format %{ "LI $dst, $src \t// compressed ptr" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addi);
__ li($dst$$Register, 0);
%}
ins_pipe(pipe_class_default);
%}
// Load hi part of compressed oop constant.
instruct loadConN_hi(iRegNdst dst, immN src) %{
effect(DEF dst, USE src);
ins_cost(DEFAULT_COST);
format %{ "LIS $dst, $src \t// narrow oop hi" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addis);
__ lis($dst$$Register, (int)(short)(($src$$constant >> 16) & 0xffff));
%}
ins_pipe(pipe_class_default);
%}
// Add lo part of compressed oop constant to already loaded hi part.
instruct loadConN_lo(iRegNdst dst, iRegNsrc src1, immN src2) %{
effect(DEF dst, USE src1, USE src2);
ins_cost(DEFAULT_COST);
format %{ "ORI $dst, $src1, $src2 \t// narrow oop lo" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addi);
assert(__ oop_recorder() != NULL, "this assembler needs an OopRecorder");
int oop_index = __ oop_recorder()->find_index((jobject)$src2$$constant);
RelocationHolder rspec = oop_Relocation::spec(oop_index);
__ relocate(rspec, 1);
__ ori($dst$$Register, $src1$$Register, $src2$$constant & 0xffff);
%}
ins_pipe(pipe_class_default);
%}
// Needed to postalloc expand loadConN: ConN is loaded as ConI
// leaving the upper 32 bits with sign-extension bits.
// This clears these bits: dst = src & 0xFFFFFFFF.
// TODO: Eventually call this maskN_regN_FFFFFFFF.
instruct clearMs32b(iRegNdst dst, iRegNsrc src) %{
effect(DEF dst, USE src);
predicate(false);
format %{ "MASK $dst, $src, 0xFFFFFFFF" %} // mask
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicl);
__ clrldi($dst$$Register, $src$$Register, 0x20);
%}
ins_pipe(pipe_class_default);
%}
// Optimize DecodeN for disjoint base.
// Load base of compressed oops into a register
instruct loadBase(iRegLdst dst) %{
effect(DEF dst);
format %{ "LoadConst $dst, heapbase" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ load_const_optimized($dst$$Register, Universe::narrow_oop_base(), R0);
%}
ins_pipe(pipe_class_default);
%}
// Loading ConN must be postalloc expanded so that edges between
// the nodes are safe. They may not interfere with a safepoint.
// GL TODO: This needs three instructions: better put this into the constant pool.
instruct loadConN_Ex(iRegNdst dst, immN src) %{
match(Set dst src);
ins_cost(DEFAULT_COST*2);
format %{ "LoadN $dst, $src \t// postalloc expanded" %} // mask
postalloc_expand %{
MachNode *m1 = new loadConN_hiNode();
MachNode *m2 = new loadConN_loNode();
MachNode *m3 = new clearMs32bNode();
m1->add_req(NULL);
m2->add_req(NULL, m1);
m3->add_req(NULL, m2);
m1->_opnds[0] = op_dst;
m1->_opnds[1] = op_src;
m2->_opnds[0] = op_dst;
m2->_opnds[1] = op_dst;
m2->_opnds[2] = op_src;
m3->_opnds[0] = op_dst;
m3->_opnds[1] = op_dst;
ra_->set_pair(m1->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
ra_->set_pair(m2->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
ra_->set_pair(m3->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
nodes->push(m1);
nodes->push(m2);
nodes->push(m3);
%}
%}
// We have seen a safepoint between the hi and lo parts, and this node was handled
// as an oop. Therefore this needs a match rule so that build_oop_map knows this is
// not a narrow oop.
instruct loadConNKlass_hi(iRegNdst dst, immNKlass_NM src) %{
match(Set dst src);
effect(DEF dst, USE src);
ins_cost(DEFAULT_COST);
format %{ "LIS $dst, $src \t// narrow klass hi" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addis);
intptr_t Csrc = Klass::encode_klass((Klass *)$src$$constant);
__ lis($dst$$Register, (int)(short)((Csrc >> 16) & 0xffff));
%}
ins_pipe(pipe_class_default);
%}
// As loadConNKlass_hi this must be recognized as narrow klass, not oop!
instruct loadConNKlass_mask(iRegNdst dst, immNKlass_NM src1, iRegNsrc src2) %{
match(Set dst src1);
effect(TEMP src2);
ins_cost(DEFAULT_COST);
format %{ "MASK $dst, $src2, 0xFFFFFFFF" %} // mask
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicl);
__ clrldi($dst$$Register, $src2$$Register, 0x20);
%}
ins_pipe(pipe_class_default);
%}
// This needs a match rule so that build_oop_map knows this is
// not a narrow oop.
instruct loadConNKlass_lo(iRegNdst dst, immNKlass_NM src1, iRegNsrc src2) %{
match(Set dst src1);
effect(TEMP src2);
ins_cost(DEFAULT_COST);
format %{ "ORI $dst, $src1, $src2 \t// narrow klass lo" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_ori);
intptr_t Csrc = Klass::encode_klass((Klass *)$src1$$constant);
assert(__ oop_recorder() != NULL, "this assembler needs an OopRecorder");
int klass_index = __ oop_recorder()->find_index((Klass *)$src1$$constant);
RelocationHolder rspec = metadata_Relocation::spec(klass_index);
__ relocate(rspec, 1);
__ ori($dst$$Register, $src2$$Register, Csrc & 0xffff);
%}
ins_pipe(pipe_class_default);
%}
// Loading ConNKlass must be postalloc expanded so that edges between
// the nodes are safe. They may not interfere with a safepoint.
instruct loadConNKlass_Ex(iRegNdst dst, immNKlass src) %{
match(Set dst src);
ins_cost(DEFAULT_COST*2);
format %{ "LoadN $dst, $src \t// postalloc expanded" %} // mask
postalloc_expand %{
// Load high bits into register. Sign extended.
MachNode *m1 = new loadConNKlass_hiNode();
m1->add_req(NULL);
m1->_opnds[0] = op_dst;
m1->_opnds[1] = op_src;
ra_->set_pair(m1->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
nodes->push(m1);
MachNode *m2 = m1;
if (!Assembler::is_uimm((jlong)Klass::encode_klass((Klass *)op_src->constant()), 31)) {
// Value might be 1-extended. Mask out these bits.
m2 = new loadConNKlass_maskNode();
m2->add_req(NULL, m1);
m2->_opnds[0] = op_dst;
m2->_opnds[1] = op_src;
m2->_opnds[2] = op_dst;
ra_->set_pair(m2->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
nodes->push(m2);
}
MachNode *m3 = new loadConNKlass_loNode();
m3->add_req(NULL, m2);
m3->_opnds[0] = op_dst;
m3->_opnds[1] = op_src;
m3->_opnds[2] = op_dst;
ra_->set_pair(m3->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
nodes->push(m3);
%}
%}
// 0x1 is used in object initialization (initial object header).
// No constant pool entries required.
instruct loadConP0or1(iRegPdst dst, immP_0or1 src) %{
match(Set dst src);
format %{ "LI $dst, $src \t// ptr" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addi);
__ li($dst$$Register, (int)((short)($src$$constant & 0xFFFF)));
%}
ins_pipe(pipe_class_default);
%}
// Expand node for constant pool load: small offset.
// The match rule is needed to generate the correct bottom_type(),
// however this node should never match. The use of predicate is not
// possible since ADLC forbids predicates for chain rules. The higher
// costs do not prevent matching in this case. For that reason the
// operand immP_NM with predicate(false) is used.
instruct loadConP(iRegPdst dst, immP_NM src, iRegLdst toc) %{
match(Set dst src);
effect(TEMP toc);
ins_num_consts(1);
format %{ "LD $dst, offset, $toc \t// load ptr $src from TOC" %}
size(4);
ins_encode( enc_load_long_constP(dst, src, toc) );
ins_pipe(pipe_class_memory);
%}
// Expand node for constant pool load: large offset.
instruct loadConP_hi(iRegPdst dst, immP_NM src, iRegLdst toc) %{
effect(DEF dst, USE src, USE toc);
predicate(false);
ins_num_consts(1);
ins_field_const_toc_offset(int);
format %{ "ADDIS $dst, $toc, offset \t// load ptr $src from TOC (hi)" %}
size(4);
ins_encode( enc_load_long_constP_hi(dst, src, toc) );
ins_pipe(pipe_class_default);
%}
// Expand node for constant pool load: large offset.
instruct loadConP_lo(iRegPdst dst, immP_NM src, iRegLdst base) %{
match(Set dst src);
effect(TEMP base);
ins_field_const_toc_offset_hi_node(loadConP_hiNode*);
format %{ "LD $dst, offset, $base \t// load ptr $src from TOC (lo)" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_ld);
int offset = ra_->C->in_scratch_emit_size() ? 0 : _const_toc_offset_hi_node->_const_toc_offset;
__ ld($dst$$Register, MacroAssembler::largeoffset_si16_si16_lo(offset), $base$$Register);
%}
ins_pipe(pipe_class_memory);
%}
// Load pointer constant from constant table. Expand in case an
// offset > 16 bit is needed.
// Adlc adds toc node MachConstantTableBase.
instruct loadConP_Ex(iRegPdst dst, immP src) %{
match(Set dst src);
ins_cost(MEMORY_REF_COST);
// This rule does not use "expand" because then
// the result type is not known to be an Oop. An ADLC
// enhancement will be needed to make that work - not worth it!
// If this instruction rematerializes, it prolongs the live range
// of the toc node, causing illegal graphs.
// assert(edge_from_to(_reg_node[reg_lo],def)) fails in verify_good_schedule().
ins_cannot_rematerialize(true);
format %{ "LD $dst, offset, $constanttablebase \t// load ptr $src from table, postalloc expanded" %}
postalloc_expand( postalloc_expand_load_ptr_constant(dst, src, constanttablebase) );
%}
// Expand node for constant pool load: small offset.
instruct loadConF(regF dst, immF src, iRegLdst toc) %{
effect(DEF dst, USE src, USE toc);
ins_cost(MEMORY_REF_COST);
ins_num_consts(1);
format %{ "LFS $dst, offset, $toc \t// load float $src from TOC" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_lfs);
address float_address = __ float_constant($src$$constant);
if (float_address == NULL) {
ciEnv::current()->record_out_of_memory_failure();
return;
}
__ lfs($dst$$FloatRegister, __ offset_to_method_toc(float_address), $toc$$Register);
%}
ins_pipe(pipe_class_memory);
%}
// Expand node for constant pool load: large offset.
instruct loadConFComp(regF dst, immF src, iRegLdst toc) %{
effect(DEF dst, USE src, USE toc);
ins_cost(MEMORY_REF_COST);
ins_num_consts(1);
format %{ "ADDIS $toc, $toc, offset_hi\n\t"
"LFS $dst, offset_lo, $toc \t// load float $src from TOC (hi/lo)\n\t"
"ADDIS $toc, $toc, -offset_hi"%}
size(12);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
FloatRegister Rdst = $dst$$FloatRegister;
Register Rtoc = $toc$$Register;
address float_address = __ float_constant($src$$constant);
if (float_address == NULL) {
ciEnv::current()->record_out_of_memory_failure();
return;
}
int offset = __ offset_to_method_toc(float_address);
int hi = (offset + (1<<15))>>16;
int lo = offset - hi * (1<<16);
__ addis(Rtoc, Rtoc, hi);
__ lfs(Rdst, lo, Rtoc);
__ addis(Rtoc, Rtoc, -hi);
%}
ins_pipe(pipe_class_memory);
%}
// Adlc adds toc node MachConstantTableBase.
instruct loadConF_Ex(regF dst, immF src) %{
match(Set dst src);
ins_cost(MEMORY_REF_COST);
// See loadConP.
ins_cannot_rematerialize(true);
format %{ "LFS $dst, offset, $constanttablebase \t// load $src from table, postalloc expanded" %}
postalloc_expand( postalloc_expand_load_float_constant(dst, src, constanttablebase) );
%}
// Expand node for constant pool load: small offset.
instruct loadConD(regD dst, immD src, iRegLdst toc) %{
effect(DEF dst, USE src, USE toc);
ins_cost(MEMORY_REF_COST);
ins_num_consts(1);
format %{ "LFD $dst, offset, $toc \t// load double $src from TOC" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_lfd);
address float_address = __ double_constant($src$$constant);
if (float_address == NULL) {
ciEnv::current()->record_out_of_memory_failure();
return;
}
int offset = __ offset_to_method_toc(float_address);
__ lfd($dst$$FloatRegister, offset, $toc$$Register);
%}
ins_pipe(pipe_class_memory);
%}
// Expand node for constant pool load: large offset.
instruct loadConDComp(regD dst, immD src, iRegLdst toc) %{
effect(DEF dst, USE src, USE toc);
ins_cost(MEMORY_REF_COST);
ins_num_consts(1);
format %{ "ADDIS $toc, $toc, offset_hi\n\t"
"LFD $dst, offset_lo, $toc \t// load double $src from TOC (hi/lo)\n\t"
"ADDIS $toc, $toc, -offset_hi" %}
size(12);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
FloatRegister Rdst = $dst$$FloatRegister;
Register Rtoc = $toc$$Register;
address float_address = __ double_constant($src$$constant);
if (float_address == NULL) {
ciEnv::current()->record_out_of_memory_failure();
return;
}
int offset = __ offset_to_method_toc(float_address);
int hi = (offset + (1<<15))>>16;
int lo = offset - hi * (1<<16);
__ addis(Rtoc, Rtoc, hi);
__ lfd(Rdst, lo, Rtoc);
__ addis(Rtoc, Rtoc, -hi);
%}
ins_pipe(pipe_class_memory);
%}
// Adlc adds toc node MachConstantTableBase.
instruct loadConD_Ex(regD dst, immD src) %{
match(Set dst src);
ins_cost(MEMORY_REF_COST);
// See loadConP.
ins_cannot_rematerialize(true);
format %{ "ConD $dst, offset, $constanttablebase \t// load $src from table, postalloc expanded" %}
postalloc_expand( postalloc_expand_load_double_constant(dst, src, constanttablebase) );
%}
// Prefetch instructions.
// Must be safe to execute with invalid address (cannot fault).
// Special prefetch versions which use the dcbz instruction.
instruct prefetch_alloc_zero(indirectMemory mem, iRegLsrc src) %{
match(PrefetchAllocation (AddP mem src));
predicate(AllocatePrefetchStyle == 3);
ins_cost(MEMORY_REF_COST);
format %{ "PREFETCH $mem, 2, $src \t// Prefetch write-many with zero" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_dcbtst);
__ dcbz($src$$Register, $mem$$base$$Register);
%}
ins_pipe(pipe_class_memory);
%}
instruct prefetch_alloc_zero_no_offset(indirectMemory mem) %{
match(PrefetchAllocation mem);
predicate(AllocatePrefetchStyle == 3);
ins_cost(MEMORY_REF_COST);
format %{ "PREFETCH $mem, 2 \t// Prefetch write-many with zero" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_dcbtst);
__ dcbz($mem$$base$$Register);
%}
ins_pipe(pipe_class_memory);
%}
instruct prefetch_alloc(indirectMemory mem, iRegLsrc src) %{
match(PrefetchAllocation (AddP mem src));
predicate(AllocatePrefetchStyle != 3);
ins_cost(MEMORY_REF_COST);
format %{ "PREFETCH $mem, 2, $src \t// Prefetch write-many" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_dcbtst);
__ dcbtst($src$$Register, $mem$$base$$Register);
%}
ins_pipe(pipe_class_memory);
%}
instruct prefetch_alloc_no_offset(indirectMemory mem) %{
match(PrefetchAllocation mem);
predicate(AllocatePrefetchStyle != 3);
ins_cost(MEMORY_REF_COST);
format %{ "PREFETCH $mem, 2 \t// Prefetch write-many" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_dcbtst);
__ dcbtst($mem$$base$$Register);
%}
ins_pipe(pipe_class_memory);
%}
//----------Store Instructions-------------------------------------------------
// Store Byte
instruct storeB(memory mem, iRegIsrc src) %{
match(Set mem (StoreB mem src));
ins_cost(MEMORY_REF_COST);
format %{ "STB $src, $mem \t// byte" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_stb);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
__ stb($src$$Register, Idisp, $mem$$base$$Register);
%}
ins_pipe(pipe_class_memory);
%}
// Store Char/Short
instruct storeC(memory mem, iRegIsrc src) %{
match(Set mem (StoreC mem src));
ins_cost(MEMORY_REF_COST);
format %{ "STH $src, $mem \t// short" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_sth);
int Idisp = $mem$$disp + frame_slots_bias($mem$$base, ra_);
__ sth($src$$Register, Idisp, $mem$$base$$Register);
%}
ins_pipe(pipe_class_memory);
%}
// Store Integer
instruct storeI(memory mem, iRegIsrc src) %{
match(Set mem (StoreI mem src));
ins_cost(MEMORY_REF_COST);
format %{ "STW $src, $mem" %}
size(4);
ins_encode( enc_stw(src, mem) );
ins_pipe(pipe_class_memory);
%}
// ConvL2I + StoreI.
instruct storeI_convL2I(memory mem, iRegLsrc src) %{
match(Set mem (StoreI mem (ConvL2I src)));
ins_cost(MEMORY_REF_COST);
format %{ "STW l2i($src), $mem" %}
size(4);
ins_encode( enc_stw(src, mem) );
ins_pipe(pipe_class_memory);
%}
// Store Long
instruct storeL(memoryAlg4 mem, iRegLsrc src) %{
match(Set mem (StoreL mem src));
ins_cost(MEMORY_REF_COST);
format %{ "STD $src, $mem \t// long" %}
size(4);
ins_encode( enc_std(src, mem) );
ins_pipe(pipe_class_memory);
%}
// Store super word nodes.
// Store Aligned Packed Byte long register to memory
instruct storeA8B(memoryAlg4 mem, iRegLsrc src) %{
predicate(n->as_StoreVector()->memory_size() == 8);
match(Set mem (StoreVector mem src));
ins_cost(MEMORY_REF_COST);
format %{ "STD $mem, $src \t// packed8B" %}
size(4);
ins_encode( enc_std(src, mem) );
ins_pipe(pipe_class_memory);
%}
// Store Compressed Oop
instruct storeN(memory dst, iRegN_P2N src) %{
match(Set dst (StoreN dst src));
ins_cost(MEMORY_REF_COST);
format %{ "STW $src, $dst \t// compressed oop" %}
size(4);
ins_encode( enc_stw(src, dst) );
ins_pipe(pipe_class_memory);
%}
// Store Compressed KLass
instruct storeNKlass(memory dst, iRegN_P2N src) %{
match(Set dst (StoreNKlass dst src));
ins_cost(MEMORY_REF_COST);
format %{ "STW $src, $dst \t// compressed klass" %}
size(4);
ins_encode( enc_stw(src, dst) );
ins_pipe(pipe_class_memory);
%}
// Store Pointer
instruct storeP(memoryAlg4 dst, iRegPsrc src) %{
match(Set dst (StoreP dst src));
ins_cost(MEMORY_REF_COST);
format %{ "STD $src, $dst \t// ptr" %}
size(4);
ins_encode( enc_std(src, dst) );
ins_pipe(pipe_class_memory);
%}
// Store Float
instruct storeF(memory mem, regF src) %{
match(Set mem (StoreF mem src));
ins_cost(MEMORY_REF_COST);
format %{ "STFS $src, $mem" %}
size(4);
ins_encode( enc_stfs(src, mem) );
ins_pipe(pipe_class_memory);
%}
// Store Double
instruct storeD(memory mem, regD src) %{
match(Set mem (StoreD mem src));
ins_cost(MEMORY_REF_COST);
format %{ "STFD $src, $mem" %}
size(4);
ins_encode( enc_stfd(src, mem) );
ins_pipe(pipe_class_memory);
%}
//----------Store Instructions With Zeros--------------------------------------
// Card-mark for CMS garbage collection.
// This cardmark does an optimization so that it must not always
// do a releasing store. For this, it gets the address of
// CMSCollectorCardTableModRefBSExt::_requires_release as input.
// (Using releaseFieldAddr in the match rule is a hack.)
instruct storeCM_CMS(memory mem, iRegLdst releaseFieldAddr, flagsReg crx) %{
match(Set mem (StoreCM mem releaseFieldAddr));
effect(TEMP crx);
predicate(false);
ins_cost(MEMORY_REF_COST);
// See loadConP.
ins_cannot_rematerialize(true);
format %{ "STB #0, $mem \t// CMS card-mark byte (must be 0!), checking requires_release in [$releaseFieldAddr]" %}
ins_encode( enc_cms_card_mark(mem, releaseFieldAddr, crx) );
ins_pipe(pipe_class_memory);
%}
// Card-mark for CMS garbage collection.
// This cardmark does an optimization so that it must not always
// do a releasing store. For this, it needs the constant address of
// CMSCollectorCardTableModRefBSExt::_requires_release.
// This constant address is split off here by expand so we can use
// adlc / matcher functionality to load it from the constant section.
instruct storeCM_CMS_ExEx(memory mem, immI_0 zero) %{
match(Set mem (StoreCM mem zero));
predicate(UseConcMarkSweepGC);
expand %{
immL baseImm %{ 0 /* TODO: PPC port (jlong)CMSCollectorCardTableModRefBSExt::requires_release_address() */ %}
iRegLdst releaseFieldAddress;
flagsReg crx;
loadConL_Ex(releaseFieldAddress, baseImm);
storeCM_CMS(mem, releaseFieldAddress, crx);
%}
%}
instruct storeCM_G1(memory mem, immI_0 zero) %{
match(Set mem (StoreCM mem zero));
predicate(UseG1GC);
ins_cost(MEMORY_REF_COST);
ins_cannot_rematerialize(true);
format %{ "STB #0, $mem \t// CMS card-mark byte store (G1)" %}
size(8);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ li(R0, 0);
//__ release(); // G1: oops are allowed to get visible after dirty marking
guarantee($mem$$base$$Register != R1_SP, "use frame_slots_bias");
__ stb(R0, $mem$$disp, $mem$$base$$Register);
%}
ins_pipe(pipe_class_memory);
%}
// Convert oop pointer into compressed form.
// Nodes for postalloc expand.
// Shift node for expand.
instruct encodeP_shift(iRegNdst dst, iRegNsrc src) %{
// The match rule is needed to make it a 'MachTypeNode'!
match(Set dst (EncodeP src));
predicate(false);
format %{ "SRDI $dst, $src, 3 \t// encode" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicl);
__ srdi($dst$$Register, $src$$Register, Universe::narrow_oop_shift() & 0x3f);
%}
ins_pipe(pipe_class_default);
%}
// Add node for expand.
instruct encodeP_sub(iRegPdst dst, iRegPdst src) %{
// The match rule is needed to make it a 'MachTypeNode'!
match(Set dst (EncodeP src));
predicate(false);
format %{ "SUB $dst, $src, oop_base \t// encode" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ sub_const_optimized($dst$$Register, $src$$Register, Universe::narrow_oop_base(), R0);
%}
ins_pipe(pipe_class_default);
%}
// Conditional sub base.
instruct cond_sub_base(iRegNdst dst, flagsRegSrc crx, iRegPsrc src1) %{
// The match rule is needed to make it a 'MachTypeNode'!
match(Set dst (EncodeP (Binary crx src1)));
predicate(false);
format %{ "BEQ $crx, done\n\t"
"SUB $dst, $src1, heapbase \t// encode: subtract base if != NULL\n"
"done:" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
Label done;
__ beq($crx$$CondRegister, done);
__ sub_const_optimized($dst$$Register, $src1$$Register, Universe::narrow_oop_base(), R0);
__ bind(done);
%}
ins_pipe(pipe_class_default);
%}
// Power 7 can use isel instruction
instruct cond_set_0_oop(iRegNdst dst, flagsRegSrc crx, iRegPsrc src1) %{
// The match rule is needed to make it a 'MachTypeNode'!
match(Set dst (EncodeP (Binary crx src1)));
predicate(false);
format %{ "CMOVE $dst, $crx eq, 0, $src1 \t// encode: preserve 0" %}
size(4);
ins_encode %{
// This is a Power7 instruction for which no machine description exists.
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ isel_0($dst$$Register, $crx$$CondRegister, Assembler::equal, $src1$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Disjoint narrow oop base.
instruct encodeP_Disjoint(iRegNdst dst, iRegPsrc src) %{
match(Set dst (EncodeP src));
predicate(Universe::narrow_oop_base_disjoint());
format %{ "EXTRDI $dst, $src, #32, #3 \t// encode with disjoint base" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicl);
__ rldicl($dst$$Register, $src$$Register, 64-Universe::narrow_oop_shift(), 32);
%}
ins_pipe(pipe_class_default);
%}
// shift != 0, base != 0
instruct encodeP_Ex(iRegNdst dst, flagsReg crx, iRegPsrc src) %{
match(Set dst (EncodeP src));
effect(TEMP crx);
predicate(n->bottom_type()->make_ptr()->ptr() != TypePtr::NotNull &&
Universe::narrow_oop_shift() != 0 &&
Universe::narrow_oop_base_overlaps());
format %{ "EncodeP $dst, $crx, $src \t// postalloc expanded" %}
postalloc_expand( postalloc_expand_encode_oop(dst, src, crx));
%}
// shift != 0, base != 0
instruct encodeP_not_null_Ex(iRegNdst dst, iRegPsrc src) %{
match(Set dst (EncodeP src));
predicate(n->bottom_type()->make_ptr()->ptr() == TypePtr::NotNull &&
Universe::narrow_oop_shift() != 0 &&
Universe::narrow_oop_base_overlaps());
format %{ "EncodeP $dst, $src\t// $src != Null, postalloc expanded" %}
postalloc_expand( postalloc_expand_encode_oop_not_null(dst, src) );
%}
// shift != 0, base == 0
// TODO: This is the same as encodeP_shift. Merge!
instruct encodeP_not_null_base_null(iRegNdst dst, iRegPsrc src) %{
match(Set dst (EncodeP src));
predicate(Universe::narrow_oop_shift() != 0 &&
Universe::narrow_oop_base() ==0);
format %{ "SRDI $dst, $src, #3 \t// encodeP, $src != NULL" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicl);
__ srdi($dst$$Register, $src$$Register, Universe::narrow_oop_shift() & 0x3f);
%}
ins_pipe(pipe_class_default);
%}
// Compressed OOPs with narrow_oop_shift == 0.
// shift == 0, base == 0
instruct encodeP_narrow_oop_shift_0(iRegNdst dst, iRegPsrc src) %{
match(Set dst (EncodeP src));
predicate(Universe::narrow_oop_shift() == 0);
format %{ "MR $dst, $src \t// Ptr->Narrow" %}
// variable size, 0 or 4.
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_or);
__ mr_if_needed($dst$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Decode nodes.
// Shift node for expand.
instruct decodeN_shift(iRegPdst dst, iRegPsrc src) %{
// The match rule is needed to make it a 'MachTypeNode'!
match(Set dst (DecodeN src));
predicate(false);
format %{ "SLDI $dst, $src, #3 \t// DecodeN" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicr);
__ sldi($dst$$Register, $src$$Register, Universe::narrow_oop_shift());
%}
ins_pipe(pipe_class_default);
%}
// Add node for expand.
instruct decodeN_add(iRegPdst dst, iRegPdst src) %{
// The match rule is needed to make it a 'MachTypeNode'!
match(Set dst (DecodeN src));
predicate(false);
format %{ "ADD $dst, $src, heapbase \t// DecodeN, add oop base" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ add_const_optimized($dst$$Register, $src$$Register, Universe::narrow_oop_base(), R0);
%}
ins_pipe(pipe_class_default);
%}
// conditianal add base for expand
instruct cond_add_base(iRegPdst dst, flagsRegSrc crx, iRegPsrc src) %{
// The match rule is needed to make it a 'MachTypeNode'!
// NOTICE that the rule is nonsense - we just have to make sure that:
// - _matrule->_rChild->_opType == "DecodeN" (see InstructForm::captures_bottom_type() in formssel.cpp)
// - we have to match 'crx' to avoid an "illegal USE of non-input: flagsReg crx" error in ADLC.
match(Set dst (DecodeN (Binary crx src)));
predicate(false);
format %{ "BEQ $crx, done\n\t"
"ADD $dst, $src, heapbase \t// DecodeN: add oop base if $src != NULL\n"
"done:" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
Label done;
__ beq($crx$$CondRegister, done);
__ add_const_optimized($dst$$Register, $src$$Register, Universe::narrow_oop_base(), R0);
__ bind(done);
%}
ins_pipe(pipe_class_default);
%}
instruct cond_set_0_ptr(iRegPdst dst, flagsRegSrc crx, iRegPsrc src1) %{
// The match rule is needed to make it a 'MachTypeNode'!
// NOTICE that the rule is nonsense - we just have to make sure that:
// - _matrule->_rChild->_opType == "DecodeN" (see InstructForm::captures_bottom_type() in formssel.cpp)
// - we have to match 'crx' to avoid an "illegal USE of non-input: flagsReg crx" error in ADLC.
match(Set dst (DecodeN (Binary crx src1)));
predicate(false);
format %{ "CMOVE $dst, $crx eq, 0, $src1 \t// decode: preserve 0" %}
size(4);
ins_encode %{
// This is a Power7 instruction for which no machine description exists.
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ isel_0($dst$$Register, $crx$$CondRegister, Assembler::equal, $src1$$Register);
%}
ins_pipe(pipe_class_default);
%}
// shift != 0, base != 0
instruct decodeN_Ex(iRegPdst dst, iRegNsrc src, flagsReg crx) %{
match(Set dst (DecodeN src));
predicate((n->bottom_type()->is_oopptr()->ptr() != TypePtr::NotNull &&
n->bottom_type()->is_oopptr()->ptr() != TypePtr::Constant) &&
Universe::narrow_oop_shift() != 0 &&
Universe::narrow_oop_base() != 0);
ins_cost(4 * DEFAULT_COST); // Should be more expensive than decodeN_Disjoint_isel_Ex.
effect(TEMP crx);
format %{ "DecodeN $dst, $src \t// Kills $crx, postalloc expanded" %}
postalloc_expand( postalloc_expand_decode_oop(dst, src, crx) );
%}
// shift != 0, base == 0
instruct decodeN_nullBase(iRegPdst dst, iRegNsrc src) %{
match(Set dst (DecodeN src));
predicate(Universe::narrow_oop_shift() != 0 &&
Universe::narrow_oop_base() == 0);
format %{ "SLDI $dst, $src, #3 \t// DecodeN (zerobased)" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicr);
__ sldi($dst$$Register, $src$$Register, Universe::narrow_oop_shift());
%}
ins_pipe(pipe_class_default);
%}
// Optimize DecodeN for disjoint base.
// Shift narrow oop and or it into register that already contains the heap base.
// Base == dst must hold, and is assured by construction in postaloc_expand.
instruct decodeN_mergeDisjoint(iRegPdst dst, iRegNsrc src, iRegLsrc base) %{
match(Set dst (DecodeN src));
effect(TEMP base);
predicate(false);
format %{ "RLDIMI $dst, $src, shift, 32-shift \t// DecodeN (disjoint base)" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldimi);
__ rldimi($dst$$Register, $src$$Register, Universe::narrow_oop_shift(), 32-Universe::narrow_oop_shift());
%}
ins_pipe(pipe_class_default);
%}
// Optimize DecodeN for disjoint base.
// This node requires only one cycle on the critical path.
// We must postalloc_expand as we can not express use_def effects where
// the used register is L and the def'ed register P.
instruct decodeN_Disjoint_notNull_Ex(iRegPdst dst, iRegNsrc src) %{
match(Set dst (DecodeN src));
effect(TEMP_DEF dst);
predicate((n->bottom_type()->is_oopptr()->ptr() == TypePtr::NotNull ||
n->bottom_type()->is_oopptr()->ptr() == TypePtr::Constant) &&
Universe::narrow_oop_base_disjoint());
ins_cost(DEFAULT_COST);
format %{ "MOV $dst, heapbase \t\n"
"RLDIMI $dst, $src, shift, 32-shift \t// decode with disjoint base" %}
postalloc_expand %{
loadBaseNode *n1 = new loadBaseNode();
n1->add_req(NULL);
n1->_opnds[0] = op_dst;
decodeN_mergeDisjointNode *n2 = new decodeN_mergeDisjointNode();
n2->add_req(n_region, n_src, n1);
n2->_opnds[0] = op_dst;
n2->_opnds[1] = op_src;
n2->_opnds[2] = op_dst;
n2->_bottom_type = _bottom_type;
ra_->set_pair(n1->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
ra_->set_pair(n2->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
nodes->push(n1);
nodes->push(n2);
%}
%}
instruct decodeN_Disjoint_isel_Ex(iRegPdst dst, iRegNsrc src, flagsReg crx) %{
match(Set dst (DecodeN src));
effect(TEMP_DEF dst, TEMP crx);
predicate((n->bottom_type()->is_oopptr()->ptr() != TypePtr::NotNull &&
n->bottom_type()->is_oopptr()->ptr() != TypePtr::Constant) &&
Universe::narrow_oop_base_disjoint() && VM_Version::has_isel());
ins_cost(3 * DEFAULT_COST);
format %{ "DecodeN $dst, $src \t// decode with disjoint base using isel" %}
postalloc_expand %{
loadBaseNode *n1 = new loadBaseNode();
n1->add_req(NULL);
n1->_opnds[0] = op_dst;
cmpN_reg_imm0Node *n_compare = new cmpN_reg_imm0Node();
n_compare->add_req(n_region, n_src);
n_compare->_opnds[0] = op_crx;
n_compare->_opnds[1] = op_src;
n_compare->_opnds[2] = new immN_0Oper(TypeNarrowOop::NULL_PTR);
decodeN_mergeDisjointNode *n2 = new decodeN_mergeDisjointNode();
n2->add_req(n_region, n_src, n1);
n2->_opnds[0] = op_dst;
n2->_opnds[1] = op_src;
n2->_opnds[2] = op_dst;
n2->_bottom_type = _bottom_type;
cond_set_0_ptrNode *n_cond_set = new cond_set_0_ptrNode();
n_cond_set->add_req(n_region, n_compare, n2);
n_cond_set->_opnds[0] = op_dst;
n_cond_set->_opnds[1] = op_crx;
n_cond_set->_opnds[2] = op_dst;
n_cond_set->_bottom_type = _bottom_type;
assert(ra_->is_oop(this) == true, "A decodeN node must produce an oop!");
ra_->set_oop(n_cond_set, true);
ra_->set_pair(n1->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
ra_->set_pair(n_compare->_idx, ra_->get_reg_second(n_crx), ra_->get_reg_first(n_crx));
ra_->set_pair(n2->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
ra_->set_pair(n_cond_set->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
nodes->push(n1);
nodes->push(n_compare);
nodes->push(n2);
nodes->push(n_cond_set);
%}
%}
// src != 0, shift != 0, base != 0
instruct decodeN_notNull_addBase_Ex(iRegPdst dst, iRegNsrc src) %{
match(Set dst (DecodeN src));
predicate((n->bottom_type()->is_oopptr()->ptr() == TypePtr::NotNull ||
n->bottom_type()->is_oopptr()->ptr() == TypePtr::Constant) &&
Universe::narrow_oop_shift() != 0 &&
Universe::narrow_oop_base() != 0);
ins_cost(2 * DEFAULT_COST);
format %{ "DecodeN $dst, $src \t// $src != NULL, postalloc expanded" %}
postalloc_expand( postalloc_expand_decode_oop_not_null(dst, src));
%}
// Compressed OOPs with narrow_oop_shift == 0.
instruct decodeN_unscaled(iRegPdst dst, iRegNsrc src) %{
match(Set dst (DecodeN src));
predicate(Universe::narrow_oop_shift() == 0);
ins_cost(DEFAULT_COST);
format %{ "MR $dst, $src \t// DecodeN (unscaled)" %}
// variable size, 0 or 4.
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_or);
__ mr_if_needed($dst$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Convert compressed oop into int for vectors alignment masking.
instruct decodeN2I_unscaled(iRegIdst dst, iRegNsrc src) %{
match(Set dst (ConvL2I (CastP2X (DecodeN src))));
predicate(Universe::narrow_oop_shift() == 0);
ins_cost(DEFAULT_COST);
format %{ "MR $dst, $src \t// (int)DecodeN (unscaled)" %}
// variable size, 0 or 4.
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_or);
__ mr_if_needed($dst$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Convert klass pointer into compressed form.
// Nodes for postalloc expand.
// Shift node for expand.
instruct encodePKlass_shift(iRegNdst dst, iRegNsrc src) %{
// The match rule is needed to make it a 'MachTypeNode'!
match(Set dst (EncodePKlass src));
predicate(false);
format %{ "SRDI $dst, $src, 3 \t// encode" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicl);
__ srdi($dst$$Register, $src$$Register, Universe::narrow_klass_shift());
%}
ins_pipe(pipe_class_default);
%}
// Add node for expand.
instruct encodePKlass_sub_base(iRegPdst dst, iRegLsrc base, iRegPdst src) %{
// The match rule is needed to make it a 'MachTypeNode'!
match(Set dst (EncodePKlass (Binary base src)));
predicate(false);
format %{ "SUB $dst, $base, $src \t// encode" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_subf);
__ subf($dst$$Register, $base$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Disjoint narrow oop base.
instruct encodePKlass_Disjoint(iRegNdst dst, iRegPsrc src) %{
match(Set dst (EncodePKlass src));
predicate(false /* TODO: PPC port Universe::narrow_klass_base_disjoint()*/);
format %{ "EXTRDI $dst, $src, #32, #3 \t// encode with disjoint base" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicl);
__ rldicl($dst$$Register, $src$$Register, 64-Universe::narrow_klass_shift(), 32);
%}
ins_pipe(pipe_class_default);
%}
// shift != 0, base != 0
instruct encodePKlass_not_null_Ex(iRegNdst dst, iRegLsrc base, iRegPsrc src) %{
match(Set dst (EncodePKlass (Binary base src)));
predicate(false);
format %{ "EncodePKlass $dst, $src\t// $src != Null, postalloc expanded" %}
postalloc_expand %{
encodePKlass_sub_baseNode *n1 = new encodePKlass_sub_baseNode();
n1->add_req(n_region, n_base, n_src);
n1->_opnds[0] = op_dst;
n1->_opnds[1] = op_base;
n1->_opnds[2] = op_src;
n1->_bottom_type = _bottom_type;
encodePKlass_shiftNode *n2 = new encodePKlass_shiftNode();
n2->add_req(n_region, n1);
n2->_opnds[0] = op_dst;
n2->_opnds[1] = op_dst;
n2->_bottom_type = _bottom_type;
ra_->set_pair(n1->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
ra_->set_pair(n2->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
nodes->push(n1);
nodes->push(n2);
%}
%}
// shift != 0, base != 0
instruct encodePKlass_not_null_ExEx(iRegNdst dst, iRegPsrc src) %{
match(Set dst (EncodePKlass src));
//predicate(Universe::narrow_klass_shift() != 0 &&
// true /* TODO: PPC port Universe::narrow_klass_base_overlaps()*/);
//format %{ "EncodePKlass $dst, $src\t// $src != Null, postalloc expanded" %}
ins_cost(DEFAULT_COST*2); // Don't count constant.
expand %{
immL baseImm %{ (jlong)(intptr_t)Universe::narrow_klass_base() %}
iRegLdst base;
loadConL_Ex(base, baseImm);
encodePKlass_not_null_Ex(dst, base, src);
%}
%}
// Decode nodes.
// Shift node for expand.
instruct decodeNKlass_shift(iRegPdst dst, iRegPsrc src) %{
// The match rule is needed to make it a 'MachTypeNode'!
match(Set dst (DecodeNKlass src));
predicate(false);
format %{ "SLDI $dst, $src, #3 \t// DecodeNKlass" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicr);
__ sldi($dst$$Register, $src$$Register, Universe::narrow_klass_shift());
%}
ins_pipe(pipe_class_default);
%}
// Add node for expand.
instruct decodeNKlass_add_base(iRegPdst dst, iRegLsrc base, iRegPdst src) %{
// The match rule is needed to make it a 'MachTypeNode'!
match(Set dst (DecodeNKlass (Binary base src)));
predicate(false);
format %{ "ADD $dst, $base, $src \t// DecodeNKlass, add klass base" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_add);
__ add($dst$$Register, $base$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
// src != 0, shift != 0, base != 0
instruct decodeNKlass_notNull_addBase_Ex(iRegPdst dst, iRegLsrc base, iRegNsrc src) %{
match(Set dst (DecodeNKlass (Binary base src)));
//effect(kill src); // We need a register for the immediate result after shifting.
predicate(false);
format %{ "DecodeNKlass $dst = $base + ($src << 3) \t// $src != NULL, postalloc expanded" %}
postalloc_expand %{
decodeNKlass_add_baseNode *n1 = new decodeNKlass_add_baseNode();
n1->add_req(n_region, n_base, n_src);
n1->_opnds[0] = op_dst;
n1->_opnds[1] = op_base;
n1->_opnds[2] = op_src;
n1->_bottom_type = _bottom_type;
decodeNKlass_shiftNode *n2 = new decodeNKlass_shiftNode();
n2->add_req(n_region, n1);
n2->_opnds[0] = op_dst;
n2->_opnds[1] = op_dst;
n2->_bottom_type = _bottom_type;
ra_->set_pair(n1->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
ra_->set_pair(n2->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this));
nodes->push(n1);
nodes->push(n2);
%}
%}
// src != 0, shift != 0, base != 0
instruct decodeNKlass_notNull_addBase_ExEx(iRegPdst dst, iRegNsrc src) %{
match(Set dst (DecodeNKlass src));
// predicate(Universe::narrow_klass_shift() != 0 &&
// Universe::narrow_klass_base() != 0);
//format %{ "DecodeNKlass $dst, $src \t// $src != NULL, expanded" %}
ins_cost(DEFAULT_COST*2); // Don't count constant.
expand %{
// We add first, then we shift. Like this, we can get along with one register less.
// But we have to load the base pre-shifted.
immL baseImm %{ (jlong)((intptr_t)Universe::narrow_klass_base() >> Universe::narrow_klass_shift()) %}
iRegLdst base;
loadConL_Ex(base, baseImm);
decodeNKlass_notNull_addBase_Ex(dst, base, src);
%}
%}
//----------MemBar Instructions-----------------------------------------------
// Memory barrier flavors
instruct membar_acquire() %{
match(LoadFence);
ins_cost(4*MEMORY_REF_COST);
format %{ "MEMBAR-acquire" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_lwsync);
__ acquire();
%}
ins_pipe(pipe_class_default);
%}
instruct unnecessary_membar_acquire() %{
match(MemBarAcquire);
ins_cost(0);
format %{ " -- \t// redundant MEMBAR-acquire - empty" %}
size(0);
ins_encode( /*empty*/ );
ins_pipe(pipe_class_default);
%}
instruct membar_acquire_lock() %{
match(MemBarAcquireLock);
ins_cost(0);
format %{ " -- \t// redundant MEMBAR-acquire - empty (acquire as part of CAS in prior FastLock)" %}
size(0);
ins_encode( /*empty*/ );
ins_pipe(pipe_class_default);
%}
instruct membar_release() %{
match(MemBarRelease);
match(StoreFence);
ins_cost(4*MEMORY_REF_COST);
format %{ "MEMBAR-release" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_lwsync);
__ release();
%}
ins_pipe(pipe_class_default);
%}
instruct membar_storestore() %{
match(MemBarStoreStore);
ins_cost(4*MEMORY_REF_COST);
format %{ "MEMBAR-store-store" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_lwsync);
__ membar(Assembler::StoreStore);
%}
ins_pipe(pipe_class_default);
%}
instruct membar_release_lock() %{
match(MemBarReleaseLock);
ins_cost(0);
format %{ " -- \t// redundant MEMBAR-release - empty (release in FastUnlock)" %}
size(0);
ins_encode( /*empty*/ );
ins_pipe(pipe_class_default);
%}
instruct membar_volatile() %{
match(MemBarVolatile);
ins_cost(4*MEMORY_REF_COST);
format %{ "MEMBAR-volatile" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_sync);
__ fence();
%}
ins_pipe(pipe_class_default);
%}
// This optimization is wrong on PPC. The following pattern is not supported:
// MemBarVolatile
// ^ ^
// | |
// CtrlProj MemProj
// ^ ^
// | |
// | Load
// |
// MemBarVolatile
//
// The first MemBarVolatile could get optimized out! According to
// Vladimir, this pattern can not occur on Oracle platforms.
// However, it does occur on PPC64 (because of membars in
// inline_unsafe_load_store).
//
// Add this node again if we found a good solution for inline_unsafe_load_store().
// Don't forget to look at the implementation of post_store_load_barrier again,
// we did other fixes in that method.
//instruct unnecessary_membar_volatile() %{
// match(MemBarVolatile);
// predicate(Matcher::post_store_load_barrier(n));
// ins_cost(0);
//
// format %{ " -- \t// redundant MEMBAR-volatile - empty" %}
// size(0);
// ins_encode( /*empty*/ );
// ins_pipe(pipe_class_default);
//%}
instruct membar_CPUOrder() %{
match(MemBarCPUOrder);
ins_cost(0);
format %{ " -- \t// MEMBAR-CPUOrder - empty: PPC64 processors are self-consistent." %}
size(0);
ins_encode( /*empty*/ );
ins_pipe(pipe_class_default);
%}
//----------Conditional Move---------------------------------------------------
// Cmove using isel.
instruct cmovI_reg_isel(cmpOp cmp, flagsRegSrc crx, iRegIdst dst, iRegIsrc src) %{
match(Set dst (CMoveI (Binary cmp crx) (Binary dst src)));
predicate(VM_Version::has_isel());
ins_cost(DEFAULT_COST);
format %{ "CMOVE $cmp, $crx, $dst, $src\n\t" %}
size(4);
ins_encode %{
// This is a Power7 instruction for which no machine description
// exists. Anyways, the scheduler should be off on Power7.
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
int cc = $cmp$$cmpcode;
__ isel($dst$$Register, $crx$$CondRegister,
(Assembler::Condition)(cc & 3), /*invert*/((~cc) & 8), $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct cmovI_reg(cmpOp cmp, flagsRegSrc crx, iRegIdst dst, iRegIsrc src) %{
match(Set dst (CMoveI (Binary cmp crx) (Binary dst src)));
predicate(!VM_Version::has_isel());
ins_cost(DEFAULT_COST+BRANCH_COST);
ins_variable_size_depending_on_alignment(true);
format %{ "CMOVE $cmp, $crx, $dst, $src\n\t" %}
// Worst case is branch + move + stop, no stop without scheduler
size(false /* TODO: PPC PORT Compile::current()->do_hb_scheduling()*/ ? 12 : 8);
ins_encode( enc_cmove_reg(dst, crx, src, cmp) );
ins_pipe(pipe_class_default);
%}
instruct cmovI_imm(cmpOp cmp, flagsRegSrc crx, iRegIdst dst, immI16 src) %{
match(Set dst (CMoveI (Binary cmp crx) (Binary dst src)));
ins_cost(DEFAULT_COST+BRANCH_COST);
ins_variable_size_depending_on_alignment(true);
format %{ "CMOVE $cmp, $crx, $dst, $src\n\t" %}
// Worst case is branch + move + stop, no stop without scheduler
size(false /* TODO: PPC PORT Compile::current()->do_hb_scheduling()*/ ? 12 : 8);
ins_encode( enc_cmove_imm(dst, crx, src, cmp) );
ins_pipe(pipe_class_default);
%}
// Cmove using isel.
instruct cmovL_reg_isel(cmpOp cmp, flagsRegSrc crx, iRegLdst dst, iRegLsrc src) %{
match(Set dst (CMoveL (Binary cmp crx) (Binary dst src)));
predicate(VM_Version::has_isel());
ins_cost(DEFAULT_COST);
format %{ "CMOVE $cmp, $crx, $dst, $src\n\t" %}
size(4);
ins_encode %{
// This is a Power7 instruction for which no machine description
// exists. Anyways, the scheduler should be off on Power7.
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
int cc = $cmp$$cmpcode;
__ isel($dst$$Register, $crx$$CondRegister,
(Assembler::Condition)(cc & 3), /*invert*/((~cc) & 8), $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct cmovL_reg(cmpOp cmp, flagsRegSrc crx, iRegLdst dst, iRegLsrc src) %{
match(Set dst (CMoveL (Binary cmp crx) (Binary dst src)));
predicate(!VM_Version::has_isel());
ins_cost(DEFAULT_COST+BRANCH_COST);
ins_variable_size_depending_on_alignment(true);
format %{ "CMOVE $cmp, $crx, $dst, $src\n\t" %}
// Worst case is branch + move + stop, no stop without scheduler.
size(false /* TODO: PPC PORT Compile::current()->do_hb_scheduling()*/ ? 12 : 8);
ins_encode( enc_cmove_reg(dst, crx, src, cmp) );
ins_pipe(pipe_class_default);
%}
instruct cmovL_imm(cmpOp cmp, flagsRegSrc crx, iRegLdst dst, immL16 src) %{
match(Set dst (CMoveL (Binary cmp crx) (Binary dst src)));
ins_cost(DEFAULT_COST+BRANCH_COST);
ins_variable_size_depending_on_alignment(true);
format %{ "CMOVE $cmp, $crx, $dst, $src\n\t" %}
// Worst case is branch + move + stop, no stop without scheduler.
size(false /* TODO: PPC PORT Compile::current()->do_hb_scheduling()*/ ? 12 : 8);
ins_encode( enc_cmove_imm(dst, crx, src, cmp) );
ins_pipe(pipe_class_default);
%}
// Cmove using isel.
instruct cmovN_reg_isel(cmpOp cmp, flagsRegSrc crx, iRegNdst dst, iRegNsrc src) %{
match(Set dst (CMoveN (Binary cmp crx) (Binary dst src)));
predicate(VM_Version::has_isel());
ins_cost(DEFAULT_COST);
format %{ "CMOVE $cmp, $crx, $dst, $src\n\t" %}
size(4);
ins_encode %{
// This is a Power7 instruction for which no machine description
// exists. Anyways, the scheduler should be off on Power7.
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
int cc = $cmp$$cmpcode;
__ isel($dst$$Register, $crx$$CondRegister,
(Assembler::Condition)(cc & 3), /*invert*/((~cc) & 8), $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Conditional move for RegN. Only cmov(reg, reg).
instruct cmovN_reg(cmpOp cmp, flagsRegSrc crx, iRegNdst dst, iRegNsrc src) %{
match(Set dst (CMoveN (Binary cmp crx) (Binary dst src)));
predicate(!VM_Version::has_isel());
ins_cost(DEFAULT_COST+BRANCH_COST);
ins_variable_size_depending_on_alignment(true);
format %{ "CMOVE $cmp, $crx, $dst, $src\n\t" %}
// Worst case is branch + move + stop, no stop without scheduler.
size(false /* TODO: PPC PORT Compile::current()->do_hb_scheduling()*/ ? 12 : 8);
ins_encode( enc_cmove_reg(dst, crx, src, cmp) );
ins_pipe(pipe_class_default);
%}
instruct cmovN_imm(cmpOp cmp, flagsRegSrc crx, iRegNdst dst, immN_0 src) %{
match(Set dst (CMoveN (Binary cmp crx) (Binary dst src)));
ins_cost(DEFAULT_COST+BRANCH_COST);
ins_variable_size_depending_on_alignment(true);
format %{ "CMOVE $cmp, $crx, $dst, $src\n\t" %}
// Worst case is branch + move + stop, no stop without scheduler.
size(false /* TODO: PPC PORT Compile::current()->do_hb_scheduling()*/ ? 12 : 8);
ins_encode( enc_cmove_imm(dst, crx, src, cmp) );
ins_pipe(pipe_class_default);
%}
// Cmove using isel.
instruct cmovP_reg_isel(cmpOp cmp, flagsRegSrc crx, iRegPdst dst, iRegPsrc src) %{
match(Set dst (CMoveP (Binary cmp crx) (Binary dst src)));
predicate(VM_Version::has_isel());
ins_cost(DEFAULT_COST);
format %{ "CMOVE $cmp, $crx, $dst, $src\n\t" %}
size(4);
ins_encode %{
// This is a Power7 instruction for which no machine description
// exists. Anyways, the scheduler should be off on Power7.
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
int cc = $cmp$$cmpcode;
__ isel($dst$$Register, $crx$$CondRegister,
(Assembler::Condition)(cc & 3), /*invert*/((~cc) & 8), $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct cmovP_reg(cmpOp cmp, flagsRegSrc crx, iRegPdst dst, iRegP_N2P src) %{
match(Set dst (CMoveP (Binary cmp crx) (Binary dst src)));
predicate(!VM_Version::has_isel());
ins_cost(DEFAULT_COST+BRANCH_COST);
ins_variable_size_depending_on_alignment(true);
format %{ "CMOVE $cmp, $crx, $dst, $src\n\t" %}
// Worst case is branch + move + stop, no stop without scheduler.
size(false /* TODO: PPC PORT Compile::current()->do_hb_scheduling()*/ ? 12 : 8);
ins_encode( enc_cmove_reg(dst, crx, src, cmp) );
ins_pipe(pipe_class_default);
%}
instruct cmovP_imm(cmpOp cmp, flagsRegSrc crx, iRegPdst dst, immP_0 src) %{
match(Set dst (CMoveP (Binary cmp crx) (Binary dst src)));
ins_cost(DEFAULT_COST+BRANCH_COST);
ins_variable_size_depending_on_alignment(true);
format %{ "CMOVE $cmp, $crx, $dst, $src\n\t" %}
// Worst case is branch + move + stop, no stop without scheduler.
size(false /* TODO: PPC PORT Compile::current()->do_hb_scheduling()*/ ? 12 : 8);
ins_encode( enc_cmove_imm(dst, crx, src, cmp) );
ins_pipe(pipe_class_default);
%}
instruct cmovF_reg(cmpOp cmp, flagsRegSrc crx, regF dst, regF src) %{
match(Set dst (CMoveF (Binary cmp crx) (Binary dst src)));
ins_cost(DEFAULT_COST+BRANCH_COST);
ins_variable_size_depending_on_alignment(true);
format %{ "CMOVEF $cmp, $crx, $dst, $src\n\t" %}
// Worst case is branch + move + stop, no stop without scheduler.
size(false /* TODO: PPC PORT (InsertEndGroupPPC64 && Compile::current()->do_hb_scheduling())*/ ? 12 : 8);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_cmovef);
Label done;
assert((Assembler::bcondCRbiIs1 & ~Assembler::bcondCRbiIs0) == 8, "check encoding");
// Branch if not (cmp crx).
__ bc(cc_to_inverse_boint($cmp$$cmpcode), cc_to_biint($cmp$$cmpcode, $crx$$reg), done);
__ fmr($dst$$FloatRegister, $src$$FloatRegister);
// TODO PPC port __ endgroup_if_needed(_size == 12);
__ bind(done);
%}
ins_pipe(pipe_class_default);
%}
instruct cmovD_reg(cmpOp cmp, flagsRegSrc crx, regD dst, regD src) %{
match(Set dst (CMoveD (Binary cmp crx) (Binary dst src)));
ins_cost(DEFAULT_COST+BRANCH_COST);
ins_variable_size_depending_on_alignment(true);
format %{ "CMOVEF $cmp, $crx, $dst, $src\n\t" %}
// Worst case is branch + move + stop, no stop without scheduler.
size(false /* TODO: PPC PORT (InsertEndGroupPPC64 && Compile::current()->do_hb_scheduling())*/ ? 12 : 8);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_cmovef);
Label done;
assert((Assembler::bcondCRbiIs1 & ~Assembler::bcondCRbiIs0) == 8, "check encoding");
// Branch if not (cmp crx).
__ bc(cc_to_inverse_boint($cmp$$cmpcode), cc_to_biint($cmp$$cmpcode, $crx$$reg), done);
__ fmr($dst$$FloatRegister, $src$$FloatRegister);
// TODO PPC port __ endgroup_if_needed(_size == 12);
__ bind(done);
%}
ins_pipe(pipe_class_default);
%}
//----------Conditional_store--------------------------------------------------
// Conditional-store of the updated heap-top.
// Used during allocation of the shared heap.
// Sets flags (EQ) on success. Implemented with a CASA on Sparc.
// As compareAndSwapL, but return flag register instead of boolean value in
// int register.
// Used by sun/misc/AtomicLongCSImpl.java.
// Mem_ptr must be a memory operand, else this node does not get
// Flag_needs_anti_dependence_check set by adlc. If this is not set this node
// can be rematerialized which leads to errors.
instruct storeLConditional_regP_regL_regL(flagsReg crx, indirect mem_ptr, iRegLsrc oldVal, iRegLsrc newVal, flagsRegCR0 cr0) %{
match(Set crx (StoreLConditional mem_ptr (Binary oldVal newVal)));
effect(TEMP cr0);
format %{ "CMPXCHGD if ($crx = ($oldVal == *$mem_ptr)) *mem_ptr = $newVal; as bool" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ cmpxchgd($crx$$CondRegister, R0, $oldVal$$Register, $newVal$$Register, $mem_ptr$$Register,
MacroAssembler::MemBarAcq, MacroAssembler::cmpxchgx_hint_atomic_update(),
noreg, NULL, true);
%}
ins_pipe(pipe_class_default);
%}
// As compareAndSwapP, but return flag register instead of boolean value in
// int register.
// This instruction is matched if UseTLAB is off.
// Mem_ptr must be a memory operand, else this node does not get
// Flag_needs_anti_dependence_check set by adlc. If this is not set this node
// can be rematerialized which leads to errors.
instruct storePConditional_regP_regP_regP(flagsRegCR0 cr0, indirect mem_ptr, iRegPsrc oldVal, iRegPsrc newVal) %{
match(Set cr0 (StorePConditional mem_ptr (Binary oldVal newVal)));
ins_cost(2*MEMORY_REF_COST);
format %{ "STDCX_ if ($cr0 = ($oldVal == *$mem_ptr)) *mem_ptr = $newVal; as bool" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_stdcx_);
__ stdcx_($newVal$$Register, $mem_ptr$$Register);
%}
ins_pipe(pipe_class_memory);
%}
// Implement LoadPLocked. Must be ordered against changes of the memory location
// by storePConditional.
// Don't know whether this is ever used.
instruct loadPLocked(iRegPdst dst, memory mem) %{
match(Set dst (LoadPLocked mem));
ins_cost(2*MEMORY_REF_COST);
format %{ "LDARX $dst, $mem \t// loadPLocked\n\t" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_ldarx);
__ ldarx($dst$$Register, $mem$$Register, MacroAssembler::cmpxchgx_hint_atomic_update());
%}
ins_pipe(pipe_class_memory);
%}
//----------Compare-And-Swap---------------------------------------------------
// CompareAndSwap{P,I,L} have more than one output, therefore "CmpI
// (CompareAndSwap ...)" or "If (CmpI (CompareAndSwap ..))" cannot be
// matched.
instruct compareAndSwapI_regP_regI_regI(iRegIdst res, iRegPdst mem_ptr, iRegIsrc src1, iRegIsrc src2, flagsRegCR0 cr0) %{
match(Set res (CompareAndSwapI mem_ptr (Binary src1 src2)));
effect(TEMP cr0);
format %{ "CMPXCHGW $res, $mem_ptr, $src1, $src2; as bool" %}
// Variable size: instruction count smaller if regs are disjoint.
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
// CmpxchgX sets CCR0 to cmpX(src1, src2) and Rres to 'true'/'false'.
__ cmpxchgw(CCR0, R0, $src1$$Register, $src2$$Register, $mem_ptr$$Register,
MacroAssembler::MemBarFenceAfter, MacroAssembler::cmpxchgx_hint_atomic_update(),
$res$$Register, true);
%}
ins_pipe(pipe_class_default);
%}
instruct compareAndSwapN_regP_regN_regN(iRegIdst res, iRegPdst mem_ptr, iRegNsrc src1, iRegNsrc src2, flagsRegCR0 cr0) %{
match(Set res (CompareAndSwapN mem_ptr (Binary src1 src2)));
effect(TEMP cr0);
format %{ "CMPXCHGW $res, $mem_ptr, $src1, $src2; as bool" %}
// Variable size: instruction count smaller if regs are disjoint.
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
// CmpxchgX sets CCR0 to cmpX(src1, src2) and Rres to 'true'/'false'.
__ cmpxchgw(CCR0, R0, $src1$$Register, $src2$$Register, $mem_ptr$$Register,
MacroAssembler::MemBarFenceAfter, MacroAssembler::cmpxchgx_hint_atomic_update(),
$res$$Register, true);
%}
ins_pipe(pipe_class_default);
%}
instruct compareAndSwapL_regP_regL_regL(iRegIdst res, iRegPdst mem_ptr, iRegLsrc src1, iRegLsrc src2, flagsRegCR0 cr0) %{
match(Set res (CompareAndSwapL mem_ptr (Binary src1 src2)));
effect(TEMP cr0);
format %{ "CMPXCHGD $res, $mem_ptr, $src1, $src2; as bool" %}
// Variable size: instruction count smaller if regs are disjoint.
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
// CmpxchgX sets CCR0 to cmpX(src1, src2) and Rres to 'true'/'false'.
__ cmpxchgd(CCR0, R0, $src1$$Register, $src2$$Register, $mem_ptr$$Register,
MacroAssembler::MemBarFenceAfter, MacroAssembler::cmpxchgx_hint_atomic_update(),
$res$$Register, NULL, true);
%}
ins_pipe(pipe_class_default);
%}
instruct compareAndSwapP_regP_regP_regP(iRegIdst res, iRegPdst mem_ptr, iRegPsrc src1, iRegPsrc src2, flagsRegCR0 cr0) %{
match(Set res (CompareAndSwapP mem_ptr (Binary src1 src2)));
effect(TEMP cr0);
format %{ "CMPXCHGD $res, $mem_ptr, $src1, $src2; as bool; ptr" %}
// Variable size: instruction count smaller if regs are disjoint.
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
// CmpxchgX sets CCR0 to cmpX(src1, src2) and Rres to 'true'/'false'.
__ cmpxchgd(CCR0, R0, $src1$$Register, $src2$$Register, $mem_ptr$$Register,
MacroAssembler::MemBarFenceAfter, MacroAssembler::cmpxchgx_hint_atomic_update(),
$res$$Register, NULL, true);
%}
ins_pipe(pipe_class_default);
%}
instruct getAndAddI(iRegIdst res, iRegPdst mem_ptr, iRegIsrc src, flagsRegCR0 cr0) %{
match(Set res (GetAndAddI mem_ptr src));
effect(TEMP cr0);
format %{ "GetAndAddI $res, $mem_ptr, $src" %}
// Variable size: instruction count smaller if regs are disjoint.
ins_encode( enc_GetAndAddI(res, mem_ptr, src) );
ins_pipe(pipe_class_default);
%}
instruct getAndAddL(iRegLdst res, iRegPdst mem_ptr, iRegLsrc src, flagsRegCR0 cr0) %{
match(Set res (GetAndAddL mem_ptr src));
effect(TEMP cr0);
format %{ "GetAndAddL $res, $mem_ptr, $src" %}
// Variable size: instruction count smaller if regs are disjoint.
ins_encode( enc_GetAndAddL(res, mem_ptr, src) );
ins_pipe(pipe_class_default);
%}
instruct getAndSetI(iRegIdst res, iRegPdst mem_ptr, iRegIsrc src, flagsRegCR0 cr0) %{
match(Set res (GetAndSetI mem_ptr src));
effect(TEMP cr0);
format %{ "GetAndSetI $res, $mem_ptr, $src" %}
// Variable size: instruction count smaller if regs are disjoint.
ins_encode( enc_GetAndSetI(res, mem_ptr, src) );
ins_pipe(pipe_class_default);
%}
instruct getAndSetL(iRegLdst res, iRegPdst mem_ptr, iRegLsrc src, flagsRegCR0 cr0) %{
match(Set res (GetAndSetL mem_ptr src));
effect(TEMP cr0);
format %{ "GetAndSetL $res, $mem_ptr, $src" %}
// Variable size: instruction count smaller if regs are disjoint.
ins_encode( enc_GetAndSetL(res, mem_ptr, src) );
ins_pipe(pipe_class_default);
%}
instruct getAndSetP(iRegPdst res, iRegPdst mem_ptr, iRegPsrc src, flagsRegCR0 cr0) %{
match(Set res (GetAndSetP mem_ptr src));
effect(TEMP cr0);
format %{ "GetAndSetP $res, $mem_ptr, $src" %}
// Variable size: instruction count smaller if regs are disjoint.
ins_encode( enc_GetAndSetL(res, mem_ptr, src) );
ins_pipe(pipe_class_default);
%}
instruct getAndSetN(iRegNdst res, iRegPdst mem_ptr, iRegNsrc src, flagsRegCR0 cr0) %{
match(Set res (GetAndSetN mem_ptr src));
effect(TEMP cr0);
format %{ "GetAndSetN $res, $mem_ptr, $src" %}
// Variable size: instruction count smaller if regs are disjoint.
ins_encode( enc_GetAndSetI(res, mem_ptr, src) );
ins_pipe(pipe_class_default);
%}
//----------Arithmetic Instructions--------------------------------------------
// Addition Instructions
// Register Addition
instruct addI_reg_reg(iRegIdst dst, iRegIsrc_iRegL2Isrc src1, iRegIsrc_iRegL2Isrc src2) %{
match(Set dst (AddI src1 src2));
format %{ "ADD $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_add);
__ add($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Expand does not work with above instruct. (??)
instruct addI_reg_reg_2(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
// no match-rule
effect(DEF dst, USE src1, USE src2);
format %{ "ADD $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_add);
__ add($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct tree_addI_addI_addI_reg_reg_Ex(iRegIdst dst, iRegIsrc src1, iRegIsrc src2, iRegIsrc src3, iRegIsrc src4) %{
match(Set dst (AddI (AddI (AddI src1 src2) src3) src4));
ins_cost(DEFAULT_COST*3);
expand %{
// FIXME: we should do this in the ideal world.
iRegIdst tmp1;
iRegIdst tmp2;
addI_reg_reg(tmp1, src1, src2);
addI_reg_reg_2(tmp2, src3, src4); // Adlc complains about addI_reg_reg.
addI_reg_reg(dst, tmp1, tmp2);
%}
%}
// Immediate Addition
instruct addI_reg_imm16(iRegIdst dst, iRegIsrc src1, immI16 src2) %{
match(Set dst (AddI src1 src2));
format %{ "ADDI $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addi);
__ addi($dst$$Register, $src1$$Register, $src2$$constant);
%}
ins_pipe(pipe_class_default);
%}
// Immediate Addition with 16-bit shifted operand
instruct addI_reg_immhi16(iRegIdst dst, iRegIsrc src1, immIhi16 src2) %{
match(Set dst (AddI src1 src2));
format %{ "ADDIS $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addis);
__ addis($dst$$Register, $src1$$Register, ($src2$$constant)>>16);
%}
ins_pipe(pipe_class_default);
%}
// Long Addition
instruct addL_reg_reg(iRegLdst dst, iRegLsrc src1, iRegLsrc src2) %{
match(Set dst (AddL src1 src2));
format %{ "ADD $dst, $src1, $src2 \t// long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_add);
__ add($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Expand does not work with above instruct. (??)
instruct addL_reg_reg_2(iRegLdst dst, iRegLsrc src1, iRegLsrc src2) %{
// no match-rule
effect(DEF dst, USE src1, USE src2);
format %{ "ADD $dst, $src1, $src2 \t// long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_add);
__ add($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct tree_addL_addL_addL_reg_reg_Ex(iRegLdst dst, iRegLsrc src1, iRegLsrc src2, iRegLsrc src3, iRegLsrc src4) %{
match(Set dst (AddL (AddL (AddL src1 src2) src3) src4));
ins_cost(DEFAULT_COST*3);
expand %{
// FIXME: we should do this in the ideal world.
iRegLdst tmp1;
iRegLdst tmp2;
addL_reg_reg(tmp1, src1, src2);
addL_reg_reg_2(tmp2, src3, src4); // Adlc complains about orI_reg_reg.
addL_reg_reg(dst, tmp1, tmp2);
%}
%}
// AddL + ConvL2I.
instruct addI_regL_regL(iRegIdst dst, iRegLsrc src1, iRegLsrc src2) %{
match(Set dst (ConvL2I (AddL src1 src2)));
format %{ "ADD $dst, $src1, $src2 \t// long + l2i" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_add);
__ add($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// No constant pool entries required.
instruct addL_reg_imm16(iRegLdst dst, iRegLsrc src1, immL16 src2) %{
match(Set dst (AddL src1 src2));
format %{ "ADDI $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addi);
__ addi($dst$$Register, $src1$$Register, $src2$$constant);
%}
ins_pipe(pipe_class_default);
%}
// Long Immediate Addition with 16-bit shifted operand.
// No constant pool entries required.
instruct addL_reg_immhi16(iRegLdst dst, iRegLsrc src1, immL32hi16 src2) %{
match(Set dst (AddL src1 src2));
format %{ "ADDIS $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addis);
__ addis($dst$$Register, $src1$$Register, ($src2$$constant)>>16);
%}
ins_pipe(pipe_class_default);
%}
// Pointer Register Addition
instruct addP_reg_reg(iRegPdst dst, iRegP_N2P src1, iRegLsrc src2) %{
match(Set dst (AddP src1 src2));
format %{ "ADD $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_add);
__ add($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Pointer Immediate Addition
// No constant pool entries required.
instruct addP_reg_imm16(iRegPdst dst, iRegP_N2P src1, immL16 src2) %{
match(Set dst (AddP src1 src2));
format %{ "ADDI $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addi);
__ addi($dst$$Register, $src1$$Register, $src2$$constant);
%}
ins_pipe(pipe_class_default);
%}
// Pointer Immediate Addition with 16-bit shifted operand.
// No constant pool entries required.
instruct addP_reg_immhi16(iRegPdst dst, iRegP_N2P src1, immL32hi16 src2) %{
match(Set dst (AddP src1 src2));
format %{ "ADDIS $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addis);
__ addis($dst$$Register, $src1$$Register, ($src2$$constant)>>16);
%}
ins_pipe(pipe_class_default);
%}
//---------------------
// Subtraction Instructions
// Register Subtraction
instruct subI_reg_reg(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
match(Set dst (SubI src1 src2));
format %{ "SUBF $dst, $src2, $src1" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_subf);
__ subf($dst$$Register, $src2$$Register, $src1$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Immediate Subtraction
// Immediate Subtraction: The compiler converts "x-c0" into "x+ -c0" (see SubLNode::Ideal),
// Don't try to use addi with - $src2$$constant since it can overflow when $src2$$constant == minI16.
// SubI from constant (using subfic).
instruct subI_imm16_reg(iRegIdst dst, immI16 src1, iRegIsrc src2) %{
match(Set dst (SubI src1 src2));
format %{ "SUBI $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_subfic);
__ subfic($dst$$Register, $src2$$Register, $src1$$constant);
%}
ins_pipe(pipe_class_default);
%}
// Turn the sign-bit of an integer into a 32-bit mask, 0x0...0 for
// positive integers and 0xF...F for negative ones.
instruct signmask32I_regI(iRegIdst dst, iRegIsrc src) %{
// no match-rule, false predicate
effect(DEF dst, USE src);
predicate(false);
format %{ "SRAWI $dst, $src, #31" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_srawi);
__ srawi($dst$$Register, $src$$Register, 0x1f);
%}
ins_pipe(pipe_class_default);
%}
instruct absI_reg_Ex(iRegIdst dst, iRegIsrc src) %{
match(Set dst (AbsI src));
ins_cost(DEFAULT_COST*3);
expand %{
iRegIdst tmp1;
iRegIdst tmp2;
signmask32I_regI(tmp1, src);
xorI_reg_reg(tmp2, tmp1, src);
subI_reg_reg(dst, tmp2, tmp1);
%}
%}
instruct negI_regI(iRegIdst dst, immI_0 zero, iRegIsrc src2) %{
match(Set dst (SubI zero src2));
format %{ "NEG $dst, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_neg);
__ neg($dst$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Long subtraction
instruct subL_reg_reg(iRegLdst dst, iRegLsrc src1, iRegLsrc src2) %{
match(Set dst (SubL src1 src2));
format %{ "SUBF $dst, $src2, $src1 \t// long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_subf);
__ subf($dst$$Register, $src2$$Register, $src1$$Register);
%}
ins_pipe(pipe_class_default);
%}
// SubL + convL2I.
instruct subI_regL_regL(iRegIdst dst, iRegLsrc src1, iRegLsrc src2) %{
match(Set dst (ConvL2I (SubL src1 src2)));
format %{ "SUBF $dst, $src2, $src1 \t// long + l2i" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_subf);
__ subf($dst$$Register, $src2$$Register, $src1$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Turn the sign-bit of a long into a 64-bit mask, 0x0...0 for
// positive longs and 0xF...F for negative ones.
instruct signmask64I_regL(iRegIdst dst, iRegLsrc src) %{
// no match-rule, false predicate
effect(DEF dst, USE src);
predicate(false);
format %{ "SRADI $dst, $src, #63" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_sradi);
__ sradi($dst$$Register, $src$$Register, 0x3f);
%}
ins_pipe(pipe_class_default);
%}
// Turn the sign-bit of a long into a 64-bit mask, 0x0...0 for
// positive longs and 0xF...F for negative ones.
instruct signmask64L_regL(iRegLdst dst, iRegLsrc src) %{
// no match-rule, false predicate
effect(DEF dst, USE src);
predicate(false);
format %{ "SRADI $dst, $src, #63" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_sradi);
__ sradi($dst$$Register, $src$$Register, 0x3f);
%}
ins_pipe(pipe_class_default);
%}
// Long negation
instruct negL_reg_reg(iRegLdst dst, immL_0 zero, iRegLsrc src2) %{
match(Set dst (SubL zero src2));
format %{ "NEG $dst, $src2 \t// long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_neg);
__ neg($dst$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// NegL + ConvL2I.
instruct negI_con0_regL(iRegIdst dst, immL_0 zero, iRegLsrc src2) %{
match(Set dst (ConvL2I (SubL zero src2)));
format %{ "NEG $dst, $src2 \t// long + l2i" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_neg);
__ neg($dst$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Multiplication Instructions
// Integer Multiplication
// Register Multiplication
instruct mulI_reg_reg(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
match(Set dst (MulI src1 src2));
ins_cost(DEFAULT_COST);
format %{ "MULLW $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_mullw);
__ mullw($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Immediate Multiplication
instruct mulI_reg_imm16(iRegIdst dst, iRegIsrc src1, immI16 src2) %{
match(Set dst (MulI src1 src2));
ins_cost(DEFAULT_COST);
format %{ "MULLI $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_mulli);
__ mulli($dst$$Register, $src1$$Register, $src2$$constant);
%}
ins_pipe(pipe_class_default);
%}
instruct mulL_reg_reg(iRegLdst dst, iRegLsrc src1, iRegLsrc src2) %{
match(Set dst (MulL src1 src2));
ins_cost(DEFAULT_COST);
format %{ "MULLD $dst $src1, $src2 \t// long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_mulld);
__ mulld($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Multiply high for optimized long division by constant.
instruct mulHighL_reg_reg(iRegLdst dst, iRegLsrc src1, iRegLsrc src2) %{
match(Set dst (MulHiL src1 src2));
ins_cost(DEFAULT_COST);
format %{ "MULHD $dst $src1, $src2 \t// long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_mulhd);
__ mulhd($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Immediate Multiplication
instruct mulL_reg_imm16(iRegLdst dst, iRegLsrc src1, immL16 src2) %{
match(Set dst (MulL src1 src2));
ins_cost(DEFAULT_COST);
format %{ "MULLI $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_mulli);
__ mulli($dst$$Register, $src1$$Register, $src2$$constant);
%}
ins_pipe(pipe_class_default);
%}
// Integer Division with Immediate -1: Negate.
instruct divI_reg_immIvalueMinus1(iRegIdst dst, iRegIsrc src1, immI_minus1 src2) %{
match(Set dst (DivI src1 src2));
ins_cost(DEFAULT_COST);
format %{ "NEG $dst, $src1 \t// /-1" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_neg);
__ neg($dst$$Register, $src1$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Integer Division with constant, but not -1.
// We should be able to improve this by checking the type of src2.
// It might well be that src2 is known to be positive.
instruct divI_reg_regnotMinus1(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
match(Set dst (DivI src1 src2));
predicate(n->in(2)->find_int_con(-1) != -1); // src2 is a constant, but not -1
ins_cost(2*DEFAULT_COST);
format %{ "DIVW $dst, $src1, $src2 \t// /not-1" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_divw);
__ divw($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct cmovI_bne_negI_reg(iRegIdst dst, flagsRegSrc crx, iRegIsrc src1) %{
effect(USE_DEF dst, USE src1, USE crx);
predicate(false);
ins_variable_size_depending_on_alignment(true);
format %{ "CMOVE $dst, neg($src1), $crx" %}
// Worst case is branch + move + stop, no stop without scheduler.
size(false /* TODO: PPC PORT (InsertEndGroupPPC64 && Compile::current()->do_hb_scheduling())*/ ? 12 : 8);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_cmove);
Label done;
__ bne($crx$$CondRegister, done);
__ neg($dst$$Register, $src1$$Register);
// TODO PPC port __ endgroup_if_needed(_size == 12);
__ bind(done);
%}
ins_pipe(pipe_class_default);
%}
// Integer Division with Registers not containing constants.
instruct divI_reg_reg_Ex(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
match(Set dst (DivI src1 src2));
ins_cost(10*DEFAULT_COST);
expand %{
immI16 imm %{ (int)-1 %}
flagsReg tmp1;
cmpI_reg_imm16(tmp1, src2, imm); // check src2 == -1
divI_reg_regnotMinus1(dst, src1, src2); // dst = src1 / src2
cmovI_bne_negI_reg(dst, tmp1, src1); // cmove dst = neg(src1) if src2 == -1
%}
%}
// Long Division with Immediate -1: Negate.
instruct divL_reg_immLvalueMinus1(iRegLdst dst, iRegLsrc src1, immL_minus1 src2) %{
match(Set dst (DivL src1 src2));
ins_cost(DEFAULT_COST);
format %{ "NEG $dst, $src1 \t// /-1, long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_neg);
__ neg($dst$$Register, $src1$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Long Division with constant, but not -1.
instruct divL_reg_regnotMinus1(iRegLdst dst, iRegLsrc src1, iRegLsrc src2) %{
match(Set dst (DivL src1 src2));
predicate(n->in(2)->find_long_con(-1L) != -1L); // Src2 is a constant, but not -1.
ins_cost(2*DEFAULT_COST);
format %{ "DIVD $dst, $src1, $src2 \t// /not-1, long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_divd);
__ divd($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct cmovL_bne_negL_reg(iRegLdst dst, flagsRegSrc crx, iRegLsrc src1) %{
effect(USE_DEF dst, USE src1, USE crx);
predicate(false);
ins_variable_size_depending_on_alignment(true);
format %{ "CMOVE $dst, neg($src1), $crx" %}
// Worst case is branch + move + stop, no stop without scheduler.
size(false /* TODO: PPC PORT (InsertEndGroupPPC64 && Compile::current()->do_hb_scheduling())*/ ? 12 : 8);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_cmove);
Label done;
__ bne($crx$$CondRegister, done);
__ neg($dst$$Register, $src1$$Register);
// TODO PPC port __ endgroup_if_needed(_size == 12);
__ bind(done);
%}
ins_pipe(pipe_class_default);
%}
// Long Division with Registers not containing constants.
instruct divL_reg_reg_Ex(iRegLdst dst, iRegLsrc src1, iRegLsrc src2) %{
match(Set dst (DivL src1 src2));
ins_cost(10*DEFAULT_COST);
expand %{
immL16 imm %{ (int)-1 %}
flagsReg tmp1;
cmpL_reg_imm16(tmp1, src2, imm); // check src2 == -1
divL_reg_regnotMinus1(dst, src1, src2); // dst = src1 / src2
cmovL_bne_negL_reg(dst, tmp1, src1); // cmove dst = neg(src1) if src2 == -1
%}
%}
// Integer Remainder with registers.
instruct modI_reg_reg_Ex(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
match(Set dst (ModI src1 src2));
ins_cost(10*DEFAULT_COST);
expand %{
immI16 imm %{ (int)-1 %}
flagsReg tmp1;
iRegIdst tmp2;
iRegIdst tmp3;
cmpI_reg_imm16(tmp1, src2, imm); // check src2 == -1
divI_reg_regnotMinus1(tmp2, src1, src2); // tmp2 = src1 / src2
cmovI_bne_negI_reg(tmp2, tmp1, src1); // cmove tmp2 = neg(src1) if src2 == -1
mulI_reg_reg(tmp3, src2, tmp2); // tmp3 = src2 * tmp2
subI_reg_reg(dst, src1, tmp3); // dst = src1 - tmp3
%}
%}
// Long Remainder with registers
instruct modL_reg_reg_Ex(iRegLdst dst, iRegLsrc src1, iRegLsrc src2) %{
match(Set dst (ModL src1 src2));
ins_cost(10*DEFAULT_COST);
expand %{
immL16 imm %{ (int)-1 %}
flagsReg tmp1;
iRegLdst tmp2;
iRegLdst tmp3;
cmpL_reg_imm16(tmp1, src2, imm); // check src2 == -1
divL_reg_regnotMinus1(tmp2, src1, src2); // tmp2 = src1 / src2
cmovL_bne_negL_reg(tmp2, tmp1, src1); // cmove tmp2 = neg(src1) if src2 == -1
mulL_reg_reg(tmp3, src2, tmp2); // tmp3 = src2 * tmp2
subL_reg_reg(dst, src1, tmp3); // dst = src1 - tmp3
%}
%}
// Integer Shift Instructions
// Register Shift Left
// Clear all but the lowest #mask bits.
// Used to normalize shift amounts in registers.
instruct maskI_reg_imm(iRegIdst dst, iRegIsrc src, uimmI6 mask) %{
// no match-rule, false predicate
effect(DEF dst, USE src, USE mask);
predicate(false);
format %{ "MASK $dst, $src, $mask \t// clear $mask upper bits" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicl);
__ clrldi($dst$$Register, $src$$Register, $mask$$constant);
%}
ins_pipe(pipe_class_default);
%}
instruct lShiftI_reg_reg(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
// no match-rule, false predicate
effect(DEF dst, USE src1, USE src2);
predicate(false);
format %{ "SLW $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_slw);
__ slw($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct lShiftI_reg_reg_Ex(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
match(Set dst (LShiftI src1 src2));
ins_cost(DEFAULT_COST*2);
expand %{
uimmI6 mask %{ 0x3b /* clear 59 bits, keep 5 */ %}
iRegIdst tmpI;
maskI_reg_imm(tmpI, src2, mask);
lShiftI_reg_reg(dst, src1, tmpI);
%}
%}
// Register Shift Left Immediate
instruct lShiftI_reg_imm(iRegIdst dst, iRegIsrc src1, immI src2) %{
match(Set dst (LShiftI src1 src2));
format %{ "SLWI $dst, $src1, ($src2 & 0x1f)" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rlwinm);
__ slwi($dst$$Register, $src1$$Register, ($src2$$constant) & 0x1f);
%}
ins_pipe(pipe_class_default);
%}
// AndI with negpow2-constant + LShiftI
instruct lShiftI_andI_immInegpow2_imm5(iRegIdst dst, iRegIsrc src1, immInegpow2 src2, uimmI5 src3) %{
match(Set dst (LShiftI (AndI src1 src2) src3));
predicate(UseRotateAndMaskInstructionsPPC64);
format %{ "RLWINM $dst, lShiftI(AndI($src1, $src2), $src3)" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rlwinm); // FIXME: assert that rlwinm is equal to addi
long src2 = $src2$$constant;
long src3 = $src3$$constant;
long maskbits = src3 + log2_long((jlong) (julong) (juint) -src2);
if (maskbits >= 32) {
__ li($dst$$Register, 0); // addi
} else {
__ rlwinm($dst$$Register, $src1$$Register, src3 & 0x1f, 0, (31-maskbits) & 0x1f);
}
%}
ins_pipe(pipe_class_default);
%}
// RShiftI + AndI with negpow2-constant + LShiftI
instruct lShiftI_andI_immInegpow2_rShiftI_imm5(iRegIdst dst, iRegIsrc src1, immInegpow2 src2, uimmI5 src3) %{
match(Set dst (LShiftI (AndI (RShiftI src1 src3) src2) src3));
predicate(UseRotateAndMaskInstructionsPPC64);
format %{ "RLWINM $dst, lShiftI(AndI(RShiftI($src1, $src3), $src2), $src3)" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rlwinm); // FIXME: assert that rlwinm is equal to addi
long src2 = $src2$$constant;
long src3 = $src3$$constant;
long maskbits = src3 + log2_long((jlong) (julong) (juint) -src2);
if (maskbits >= 32) {
__ li($dst$$Register, 0); // addi
} else {
__ rlwinm($dst$$Register, $src1$$Register, 0, 0, (31-maskbits) & 0x1f);
}
%}
ins_pipe(pipe_class_default);
%}
instruct lShiftL_regL_regI(iRegLdst dst, iRegLsrc src1, iRegIsrc src2) %{
// no match-rule, false predicate
effect(DEF dst, USE src1, USE src2);
predicate(false);
format %{ "SLD $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_sld);
__ sld($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Register Shift Left
instruct lShiftL_regL_regI_Ex(iRegLdst dst, iRegLsrc src1, iRegIsrc src2) %{
match(Set dst (LShiftL src1 src2));
ins_cost(DEFAULT_COST*2);
expand %{
uimmI6 mask %{ 0x3a /* clear 58 bits, keep 6 */ %}
iRegIdst tmpI;
maskI_reg_imm(tmpI, src2, mask);
lShiftL_regL_regI(dst, src1, tmpI);
%}
%}
// Register Shift Left Immediate
instruct lshiftL_regL_immI(iRegLdst dst, iRegLsrc src1, immI src2) %{
match(Set dst (LShiftL src1 src2));
format %{ "SLDI $dst, $src1, ($src2 & 0x3f)" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicr);
__ sldi($dst$$Register, $src1$$Register, ($src2$$constant) & 0x3f);
%}
ins_pipe(pipe_class_default);
%}
// If we shift more than 32 bits, we need not convert I2L.
instruct lShiftL_regI_immGE32(iRegLdst dst, iRegIsrc src1, uimmI6_ge32 src2) %{
match(Set dst (LShiftL (ConvI2L src1) src2));
ins_cost(DEFAULT_COST);
size(4);
format %{ "SLDI $dst, i2l($src1), $src2" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicr);
__ sldi($dst$$Register, $src1$$Register, ($src2$$constant) & 0x3f);
%}
ins_pipe(pipe_class_default);
%}
// Shift a postivie int to the left.
// Clrlsldi clears the upper 32 bits and shifts.
instruct scaledPositiveI2L_lShiftL_convI2L_reg_imm6(iRegLdst dst, iRegIsrc src1, uimmI6 src2) %{
match(Set dst (LShiftL (ConvI2L src1) src2));
predicate(((ConvI2LNode*)(_kids[0]->_leaf))->type()->is_long()->is_positive_int());
format %{ "SLDI $dst, i2l(positive_int($src1)), $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldic);
__ clrlsldi($dst$$Register, $src1$$Register, 0x20, $src2$$constant);
%}
ins_pipe(pipe_class_default);
%}
instruct arShiftI_reg_reg(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
// no match-rule, false predicate
effect(DEF dst, USE src1, USE src2);
predicate(false);
format %{ "SRAW $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_sraw);
__ sraw($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Register Arithmetic Shift Right
instruct arShiftI_reg_reg_Ex(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
match(Set dst (RShiftI src1 src2));
ins_cost(DEFAULT_COST*2);
expand %{
uimmI6 mask %{ 0x3b /* clear 59 bits, keep 5 */ %}
iRegIdst tmpI;
maskI_reg_imm(tmpI, src2, mask);
arShiftI_reg_reg(dst, src1, tmpI);
%}
%}
// Register Arithmetic Shift Right Immediate
instruct arShiftI_reg_imm(iRegIdst dst, iRegIsrc src1, immI src2) %{
match(Set dst (RShiftI src1 src2));
format %{ "SRAWI $dst, $src1, ($src2 & 0x1f)" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_srawi);
__ srawi($dst$$Register, $src1$$Register, ($src2$$constant) & 0x1f);
%}
ins_pipe(pipe_class_default);
%}
instruct arShiftL_regL_regI(iRegLdst dst, iRegLsrc src1, iRegIsrc src2) %{
// no match-rule, false predicate
effect(DEF dst, USE src1, USE src2);
predicate(false);
format %{ "SRAD $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_srad);
__ srad($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Register Shift Right Arithmetic Long
instruct arShiftL_regL_regI_Ex(iRegLdst dst, iRegLsrc src1, iRegIsrc src2) %{
match(Set dst (RShiftL src1 src2));
ins_cost(DEFAULT_COST*2);
expand %{
uimmI6 mask %{ 0x3a /* clear 58 bits, keep 6 */ %}
iRegIdst tmpI;
maskI_reg_imm(tmpI, src2, mask);
arShiftL_regL_regI(dst, src1, tmpI);
%}
%}
// Register Shift Right Immediate
instruct arShiftL_regL_immI(iRegLdst dst, iRegLsrc src1, immI src2) %{
match(Set dst (RShiftL src1 src2));
format %{ "SRADI $dst, $src1, ($src2 & 0x3f)" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_sradi);
__ sradi($dst$$Register, $src1$$Register, ($src2$$constant) & 0x3f);
%}
ins_pipe(pipe_class_default);
%}
// RShiftL + ConvL2I
instruct convL2I_arShiftL_regL_immI(iRegIdst dst, iRegLsrc src1, immI src2) %{
match(Set dst (ConvL2I (RShiftL src1 src2)));
format %{ "SRADI $dst, $src1, ($src2 & 0x3f) \t// long + l2i" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_sradi);
__ sradi($dst$$Register, $src1$$Register, ($src2$$constant) & 0x3f);
%}
ins_pipe(pipe_class_default);
%}
instruct urShiftI_reg_reg(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
// no match-rule, false predicate
effect(DEF dst, USE src1, USE src2);
predicate(false);
format %{ "SRW $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_srw);
__ srw($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Register Shift Right
instruct urShiftI_reg_reg_Ex(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
match(Set dst (URShiftI src1 src2));
ins_cost(DEFAULT_COST*2);
expand %{
uimmI6 mask %{ 0x3b /* clear 59 bits, keep 5 */ %}
iRegIdst tmpI;
maskI_reg_imm(tmpI, src2, mask);
urShiftI_reg_reg(dst, src1, tmpI);
%}
%}
// Register Shift Right Immediate
instruct urShiftI_reg_imm(iRegIdst dst, iRegIsrc src1, immI src2) %{
match(Set dst (URShiftI src1 src2));
format %{ "SRWI $dst, $src1, ($src2 & 0x1f)" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rlwinm);
__ srwi($dst$$Register, $src1$$Register, ($src2$$constant) & 0x1f);
%}
ins_pipe(pipe_class_default);
%}
instruct urShiftL_regL_regI(iRegLdst dst, iRegLsrc src1, iRegIsrc src2) %{
// no match-rule, false predicate
effect(DEF dst, USE src1, USE src2);
predicate(false);
format %{ "SRD $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_srd);
__ srd($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Register Shift Right
instruct urShiftL_regL_regI_Ex(iRegLdst dst, iRegLsrc src1, iRegIsrc src2) %{
match(Set dst (URShiftL src1 src2));
ins_cost(DEFAULT_COST*2);
expand %{
uimmI6 mask %{ 0x3a /* clear 58 bits, keep 6 */ %}
iRegIdst tmpI;
maskI_reg_imm(tmpI, src2, mask);
urShiftL_regL_regI(dst, src1, tmpI);
%}
%}
// Register Shift Right Immediate
instruct urShiftL_regL_immI(iRegLdst dst, iRegLsrc src1, immI src2) %{
match(Set dst (URShiftL src1 src2));
format %{ "SRDI $dst, $src1, ($src2 & 0x3f)" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicl);
__ srdi($dst$$Register, $src1$$Register, ($src2$$constant) & 0x3f);
%}
ins_pipe(pipe_class_default);
%}
// URShiftL + ConvL2I.
instruct convL2I_urShiftL_regL_immI(iRegIdst dst, iRegLsrc src1, immI src2) %{
match(Set dst (ConvL2I (URShiftL src1 src2)));
format %{ "SRDI $dst, $src1, ($src2 & 0x3f) \t// long + l2i" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicl);
__ srdi($dst$$Register, $src1$$Register, ($src2$$constant) & 0x3f);
%}
ins_pipe(pipe_class_default);
%}
// Register Shift Right Immediate with a CastP2X
instruct shrP_convP2X_reg_imm6(iRegLdst dst, iRegP_N2P src1, uimmI6 src2) %{
match(Set dst (URShiftL (CastP2X src1) src2));
format %{ "SRDI $dst, $src1, $src2 \t// Cast ptr $src1 to long and shift" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicl);
__ srdi($dst$$Register, $src1$$Register, ($src2$$constant) & 0x3f);
%}
ins_pipe(pipe_class_default);
%}
instruct sxtI_reg(iRegIdst dst, iRegIsrc src) %{
match(Set dst (ConvL2I (ConvI2L src)));
format %{ "EXTSW $dst, $src \t// int->int" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_extsw);
__ extsw($dst$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
//----------Rotate Instructions------------------------------------------------
// Rotate Left by 8-bit immediate
instruct rotlI_reg_immi8(iRegIdst dst, iRegIsrc src, immI8 lshift, immI8 rshift) %{
match(Set dst (OrI (LShiftI src lshift) (URShiftI src rshift)));
predicate(0 == ((n->in(1)->in(2)->get_int() + n->in(2)->in(2)->get_int()) & 0x1f));
format %{ "ROTLWI $dst, $src, $lshift" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rlwinm);
__ rotlwi($dst$$Register, $src$$Register, $lshift$$constant);
%}
ins_pipe(pipe_class_default);
%}
// Rotate Right by 8-bit immediate
instruct rotrI_reg_immi8(iRegIdst dst, iRegIsrc src, immI8 rshift, immI8 lshift) %{
match(Set dst (OrI (URShiftI src rshift) (LShiftI src lshift)));
predicate(0 == ((n->in(1)->in(2)->get_int() + n->in(2)->in(2)->get_int()) & 0x1f));
format %{ "ROTRWI $dst, $rshift" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rlwinm);
__ rotrwi($dst$$Register, $src$$Register, $rshift$$constant);
%}
ins_pipe(pipe_class_default);
%}
//----------Floating Point Arithmetic Instructions-----------------------------
// Add float single precision
instruct addF_reg_reg(regF dst, regF src1, regF src2) %{
match(Set dst (AddF src1 src2));
format %{ "FADDS $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fadds);
__ fadds($dst$$FloatRegister, $src1$$FloatRegister, $src2$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
// Add float double precision
instruct addD_reg_reg(regD dst, regD src1, regD src2) %{
match(Set dst (AddD src1 src2));
format %{ "FADD $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fadd);
__ fadd($dst$$FloatRegister, $src1$$FloatRegister, $src2$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
// Sub float single precision
instruct subF_reg_reg(regF dst, regF src1, regF src2) %{
match(Set dst (SubF src1 src2));
format %{ "FSUBS $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fsubs);
__ fsubs($dst$$FloatRegister, $src1$$FloatRegister, $src2$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
// Sub float double precision
instruct subD_reg_reg(regD dst, regD src1, regD src2) %{
match(Set dst (SubD src1 src2));
format %{ "FSUB $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fsub);
__ fsub($dst$$FloatRegister, $src1$$FloatRegister, $src2$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
// Mul float single precision
instruct mulF_reg_reg(regF dst, regF src1, regF src2) %{
match(Set dst (MulF src1 src2));
format %{ "FMULS $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fmuls);
__ fmuls($dst$$FloatRegister, $src1$$FloatRegister, $src2$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
// Mul float double precision
instruct mulD_reg_reg(regD dst, regD src1, regD src2) %{
match(Set dst (MulD src1 src2));
format %{ "FMUL $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fmul);
__ fmul($dst$$FloatRegister, $src1$$FloatRegister, $src2$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
// Div float single precision
instruct divF_reg_reg(regF dst, regF src1, regF src2) %{
match(Set dst (DivF src1 src2));
format %{ "FDIVS $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fdivs);
__ fdivs($dst$$FloatRegister, $src1$$FloatRegister, $src2$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
// Div float double precision
instruct divD_reg_reg(regD dst, regD src1, regD src2) %{
match(Set dst (DivD src1 src2));
format %{ "FDIV $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fdiv);
__ fdiv($dst$$FloatRegister, $src1$$FloatRegister, $src2$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
// Absolute float single precision
instruct absF_reg(regF dst, regF src) %{
match(Set dst (AbsF src));
format %{ "FABS $dst, $src \t// float" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fabs);
__ fabs($dst$$FloatRegister, $src$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
// Absolute float double precision
instruct absD_reg(regD dst, regD src) %{
match(Set dst (AbsD src));
format %{ "FABS $dst, $src \t// double" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fabs);
__ fabs($dst$$FloatRegister, $src$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
instruct negF_reg(regF dst, regF src) %{
match(Set dst (NegF src));
format %{ "FNEG $dst, $src \t// float" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fneg);
__ fneg($dst$$FloatRegister, $src$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
instruct negD_reg(regD dst, regD src) %{
match(Set dst (NegD src));
format %{ "FNEG $dst, $src \t// double" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fneg);
__ fneg($dst$$FloatRegister, $src$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
// AbsF + NegF.
instruct negF_absF_reg(regF dst, regF src) %{
match(Set dst (NegF (AbsF src)));
format %{ "FNABS $dst, $src \t// float" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fnabs);
__ fnabs($dst$$FloatRegister, $src$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
// AbsD + NegD.
instruct negD_absD_reg(regD dst, regD src) %{
match(Set dst (NegD (AbsD src)));
format %{ "FNABS $dst, $src \t// double" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fnabs);
__ fnabs($dst$$FloatRegister, $src$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
// VM_Version::has_fsqrt() decides if this node will be used.
// Sqrt float double precision
instruct sqrtD_reg(regD dst, regD src) %{
match(Set dst (SqrtD src));
format %{ "FSQRT $dst, $src" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fsqrt);
__ fsqrt($dst$$FloatRegister, $src$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
// Single-precision sqrt.
instruct sqrtF_reg(regF dst, regF src) %{
match(Set dst (ConvD2F (SqrtD (ConvF2D src))));
predicate(VM_Version::has_fsqrts());
ins_cost(DEFAULT_COST);
format %{ "FSQRTS $dst, $src" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fsqrts);
__ fsqrts($dst$$FloatRegister, $src$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
instruct roundDouble_nop(regD dst) %{
match(Set dst (RoundDouble dst));
ins_cost(0);
format %{ " -- \t// RoundDouble not needed - empty" %}
size(0);
// PPC results are already "rounded" (i.e., normal-format IEEE).
ins_encode( /*empty*/ );
ins_pipe(pipe_class_default);
%}
instruct roundFloat_nop(regF dst) %{
match(Set dst (RoundFloat dst));
ins_cost(0);
format %{ " -- \t// RoundFloat not needed - empty" %}
size(0);
// PPC results are already "rounded" (i.e., normal-format IEEE).
ins_encode( /*empty*/ );
ins_pipe(pipe_class_default);
%}
//----------Logical Instructions-----------------------------------------------
// And Instructions
// Register And
instruct andI_reg_reg(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
match(Set dst (AndI src1 src2));
format %{ "AND $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_and);
__ andr($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Immediate And
instruct andI_reg_uimm16(iRegIdst dst, iRegIsrc src1, uimmI16 src2, flagsRegCR0 cr0) %{
match(Set dst (AndI src1 src2));
effect(KILL cr0);
format %{ "ANDI $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_andi_);
// FIXME: avoid andi_ ?
__ andi_($dst$$Register, $src1$$Register, $src2$$constant);
%}
ins_pipe(pipe_class_default);
%}
// Immediate And where the immediate is a negative power of 2.
instruct andI_reg_immInegpow2(iRegIdst dst, iRegIsrc src1, immInegpow2 src2) %{
match(Set dst (AndI src1 src2));
format %{ "ANDWI $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicr);
__ clrrdi($dst$$Register, $src1$$Register, log2_long((jlong)(julong)(juint)-($src2$$constant)));
%}
ins_pipe(pipe_class_default);
%}
instruct andI_reg_immIpow2minus1(iRegIdst dst, iRegIsrc src1, immIpow2minus1 src2) %{
match(Set dst (AndI src1 src2));
format %{ "ANDWI $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicl);
__ clrldi($dst$$Register, $src1$$Register, 64-log2_long((((jlong) $src2$$constant)+1)));
%}
ins_pipe(pipe_class_default);
%}
instruct andI_reg_immIpowerOf2(iRegIdst dst, iRegIsrc src1, immIpowerOf2 src2) %{
match(Set dst (AndI src1 src2));
predicate(UseRotateAndMaskInstructionsPPC64);
format %{ "ANDWI $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rlwinm);
__ rlwinm($dst$$Register, $src1$$Register, 0,
(31-log2_long((jlong) $src2$$constant)) & 0x1f, (31-log2_long((jlong) $src2$$constant)) & 0x1f);
%}
ins_pipe(pipe_class_default);
%}
// Register And Long
instruct andL_reg_reg(iRegLdst dst, iRegLsrc src1, iRegLsrc src2) %{
match(Set dst (AndL src1 src2));
ins_cost(DEFAULT_COST);
format %{ "AND $dst, $src1, $src2 \t// long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_and);
__ andr($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Immediate And long
instruct andL_reg_uimm16(iRegLdst dst, iRegLsrc src1, uimmL16 src2, flagsRegCR0 cr0) %{
match(Set dst (AndL src1 src2));
effect(KILL cr0);
format %{ "ANDI $dst, $src1, $src2 \t// long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_andi_);
// FIXME: avoid andi_ ?
__ andi_($dst$$Register, $src1$$Register, $src2$$constant);
%}
ins_pipe(pipe_class_default);
%}
// Immediate And Long where the immediate is a negative power of 2.
instruct andL_reg_immLnegpow2(iRegLdst dst, iRegLsrc src1, immLnegpow2 src2) %{
match(Set dst (AndL src1 src2));
format %{ "ANDDI $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicr);
__ clrrdi($dst$$Register, $src1$$Register, log2_long((jlong)-$src2$$constant));
%}
ins_pipe(pipe_class_default);
%}
instruct andL_reg_immLpow2minus1(iRegLdst dst, iRegLsrc src1, immLpow2minus1 src2) %{
match(Set dst (AndL src1 src2));
format %{ "ANDDI $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicl);
__ clrldi($dst$$Register, $src1$$Register, 64-log2_long((((jlong) $src2$$constant)+1)));
%}
ins_pipe(pipe_class_default);
%}
// AndL + ConvL2I.
instruct convL2I_andL_reg_immLpow2minus1(iRegIdst dst, iRegLsrc src1, immLpow2minus1 src2) %{
match(Set dst (ConvL2I (AndL src1 src2)));
ins_cost(DEFAULT_COST);
format %{ "ANDDI $dst, $src1, $src2 \t// long + l2i" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicl);
__ clrldi($dst$$Register, $src1$$Register, 64-log2_long((((jlong) $src2$$constant)+1)));
%}
ins_pipe(pipe_class_default);
%}
// Or Instructions
// Register Or
instruct orI_reg_reg(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
match(Set dst (OrI src1 src2));
format %{ "OR $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_or);
__ or_unchecked($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Expand does not work with above instruct. (??)
instruct orI_reg_reg_2(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
// no match-rule
effect(DEF dst, USE src1, USE src2);
format %{ "OR $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_or);
__ or_unchecked($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct tree_orI_orI_orI_reg_reg_Ex(iRegIdst dst, iRegIsrc src1, iRegIsrc src2, iRegIsrc src3, iRegIsrc src4) %{
match(Set dst (OrI (OrI (OrI src1 src2) src3) src4));
ins_cost(DEFAULT_COST*3);
expand %{
// FIXME: we should do this in the ideal world.
iRegIdst tmp1;
iRegIdst tmp2;
orI_reg_reg(tmp1, src1, src2);
orI_reg_reg_2(tmp2, src3, src4); // Adlc complains about orI_reg_reg.
orI_reg_reg(dst, tmp1, tmp2);
%}
%}
// Immediate Or
instruct orI_reg_uimm16(iRegIdst dst, iRegIsrc src1, uimmI16 src2) %{
match(Set dst (OrI src1 src2));
format %{ "ORI $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_ori);
__ ori($dst$$Register, $src1$$Register, ($src2$$constant) & 0xFFFF);
%}
ins_pipe(pipe_class_default);
%}
// Register Or Long
instruct orL_reg_reg(iRegLdst dst, iRegLsrc src1, iRegLsrc src2) %{
match(Set dst (OrL src1 src2));
ins_cost(DEFAULT_COST);
size(4);
format %{ "OR $dst, $src1, $src2 \t// long" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_or);
__ or_unchecked($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// OrL + ConvL2I.
instruct orI_regL_regL(iRegIdst dst, iRegLsrc src1, iRegLsrc src2) %{
match(Set dst (ConvL2I (OrL src1 src2)));
ins_cost(DEFAULT_COST);
format %{ "OR $dst, $src1, $src2 \t// long + l2i" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_or);
__ or_unchecked($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Immediate Or long
instruct orL_reg_uimm16(iRegLdst dst, iRegLsrc src1, uimmL16 con) %{
match(Set dst (OrL src1 con));
ins_cost(DEFAULT_COST);
format %{ "ORI $dst, $src1, $con \t// long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_ori);
__ ori($dst$$Register, $src1$$Register, ($con$$constant) & 0xFFFF);
%}
ins_pipe(pipe_class_default);
%}
// Xor Instructions
// Register Xor
instruct xorI_reg_reg(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
match(Set dst (XorI src1 src2));
format %{ "XOR $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_xor);
__ xorr($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Expand does not work with above instruct. (??)
instruct xorI_reg_reg_2(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
// no match-rule
effect(DEF dst, USE src1, USE src2);
format %{ "XOR $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_xor);
__ xorr($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct tree_xorI_xorI_xorI_reg_reg_Ex(iRegIdst dst, iRegIsrc src1, iRegIsrc src2, iRegIsrc src3, iRegIsrc src4) %{
match(Set dst (XorI (XorI (XorI src1 src2) src3) src4));
ins_cost(DEFAULT_COST*3);
expand %{
// FIXME: we should do this in the ideal world.
iRegIdst tmp1;
iRegIdst tmp2;
xorI_reg_reg(tmp1, src1, src2);
xorI_reg_reg_2(tmp2, src3, src4); // Adlc complains about xorI_reg_reg.
xorI_reg_reg(dst, tmp1, tmp2);
%}
%}
// Immediate Xor
instruct xorI_reg_uimm16(iRegIdst dst, iRegIsrc src1, uimmI16 src2) %{
match(Set dst (XorI src1 src2));
format %{ "XORI $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_xori);
__ xori($dst$$Register, $src1$$Register, $src2$$constant);
%}
ins_pipe(pipe_class_default);
%}
// Register Xor Long
instruct xorL_reg_reg(iRegLdst dst, iRegLsrc src1, iRegLsrc src2) %{
match(Set dst (XorL src1 src2));
ins_cost(DEFAULT_COST);
format %{ "XOR $dst, $src1, $src2 \t// long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_xor);
__ xorr($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// XorL + ConvL2I.
instruct xorI_regL_regL(iRegIdst dst, iRegLsrc src1, iRegLsrc src2) %{
match(Set dst (ConvL2I (XorL src1 src2)));
ins_cost(DEFAULT_COST);
format %{ "XOR $dst, $src1, $src2 \t// long + l2i" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_xor);
__ xorr($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Immediate Xor Long
instruct xorL_reg_uimm16(iRegLdst dst, iRegLsrc src1, uimmL16 src2) %{
match(Set dst (XorL src1 src2));
ins_cost(DEFAULT_COST);
format %{ "XORI $dst, $src1, $src2 \t// long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_xori);
__ xori($dst$$Register, $src1$$Register, $src2$$constant);
%}
ins_pipe(pipe_class_default);
%}
instruct notI_reg(iRegIdst dst, iRegIsrc src1, immI_minus1 src2) %{
match(Set dst (XorI src1 src2));
ins_cost(DEFAULT_COST);
format %{ "NOT $dst, $src1 ($src2)" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_nor);
__ nor($dst$$Register, $src1$$Register, $src1$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct notL_reg(iRegLdst dst, iRegLsrc src1, immL_minus1 src2) %{
match(Set dst (XorL src1 src2));
ins_cost(DEFAULT_COST);
format %{ "NOT $dst, $src1 ($src2) \t// long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_nor);
__ nor($dst$$Register, $src1$$Register, $src1$$Register);
%}
ins_pipe(pipe_class_default);
%}
// And-complement
instruct andcI_reg_reg(iRegIdst dst, iRegIsrc src1, immI_minus1 src2, iRegIsrc src3) %{
match(Set dst (AndI (XorI src1 src2) src3));
ins_cost(DEFAULT_COST);
format %{ "ANDW $dst, xori($src1, $src2), $src3" %}
size(4);
ins_encode( enc_andc(dst, src3, src1) );
ins_pipe(pipe_class_default);
%}
// And-complement
instruct andcL_reg_reg(iRegLdst dst, iRegLsrc src1, iRegLsrc src2) %{
// no match-rule, false predicate
effect(DEF dst, USE src1, USE src2);
predicate(false);
format %{ "ANDC $dst, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_andc);
__ andc($dst$$Register, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_default);
%}
//----------Moves between int/long and float/double----------------------------
//
// The following rules move values from int/long registers/stack-locations
// to float/double registers/stack-locations and vice versa, without doing any
// conversions. These rules are used to implement the bit-conversion methods
// of java.lang.Float etc., e.g.
// int floatToIntBits(float value)
// float intBitsToFloat(int bits)
//
// Notes on the implementation on ppc64:
// We only provide rules which move between a register and a stack-location,
// because we always have to go through memory when moving between a float
// register and an integer register.
//---------- Chain stack slots between similar types --------
// These are needed so that the rules below can match.
// Load integer from stack slot
instruct stkI_to_regI(iRegIdst dst, stackSlotI src) %{
match(Set dst src);
ins_cost(MEMORY_REF_COST);
format %{ "LWZ $dst, $src" %}
size(4);
ins_encode( enc_lwz(dst, src) );
ins_pipe(pipe_class_memory);
%}
// Store integer to stack slot
instruct regI_to_stkI(stackSlotI dst, iRegIsrc src) %{
match(Set dst src);
ins_cost(MEMORY_REF_COST);
format %{ "STW $src, $dst \t// stk" %}
size(4);
ins_encode( enc_stw(src, dst) ); // rs=rt
ins_pipe(pipe_class_memory);
%}
// Load long from stack slot
instruct stkL_to_regL(iRegLdst dst, stackSlotL src) %{
match(Set dst src);
ins_cost(MEMORY_REF_COST);
format %{ "LD $dst, $src \t// long" %}
size(4);
ins_encode( enc_ld(dst, src) );
ins_pipe(pipe_class_memory);
%}
// Store long to stack slot
instruct regL_to_stkL(stackSlotL dst, iRegLsrc src) %{
match(Set dst src);
ins_cost(MEMORY_REF_COST);
format %{ "STD $src, $dst \t// long" %}
size(4);
ins_encode( enc_std(src, dst) ); // rs=rt
ins_pipe(pipe_class_memory);
%}
//----------Moves between int and float
// Move float value from float stack-location to integer register.
instruct moveF2I_stack_reg(iRegIdst dst, stackSlotF src) %{
match(Set dst (MoveF2I src));
ins_cost(MEMORY_REF_COST);
format %{ "LWZ $dst, $src \t// MoveF2I" %}
size(4);
ins_encode( enc_lwz(dst, src) );
ins_pipe(pipe_class_memory);
%}
// Move float value from float register to integer stack-location.
instruct moveF2I_reg_stack(stackSlotI dst, regF src) %{
match(Set dst (MoveF2I src));
ins_cost(MEMORY_REF_COST);
format %{ "STFS $src, $dst \t// MoveF2I" %}
size(4);
ins_encode( enc_stfs(src, dst) );
ins_pipe(pipe_class_memory);
%}
// Move integer value from integer stack-location to float register.
instruct moveI2F_stack_reg(regF dst, stackSlotI src) %{
match(Set dst (MoveI2F src));
ins_cost(MEMORY_REF_COST);
format %{ "LFS $dst, $src \t// MoveI2F" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_lfs);
int Idisp = $src$$disp + frame_slots_bias($src$$base, ra_);
__ lfs($dst$$FloatRegister, Idisp, $src$$base$$Register);
%}
ins_pipe(pipe_class_memory);
%}
// Move integer value from integer register to float stack-location.
instruct moveI2F_reg_stack(stackSlotF dst, iRegIsrc src) %{
match(Set dst (MoveI2F src));
ins_cost(MEMORY_REF_COST);
format %{ "STW $src, $dst \t// MoveI2F" %}
size(4);
ins_encode( enc_stw(src, dst) );
ins_pipe(pipe_class_memory);
%}
//----------Moves between long and float
instruct moveF2L_reg_stack(stackSlotL dst, regF src) %{
// no match-rule, false predicate
effect(DEF dst, USE src);
predicate(false);
format %{ "storeD $src, $dst \t// STACK" %}
size(4);
ins_encode( enc_stfd(src, dst) );
ins_pipe(pipe_class_default);
%}
//----------Moves between long and double
// Move double value from double stack-location to long register.
instruct moveD2L_stack_reg(iRegLdst dst, stackSlotD src) %{
match(Set dst (MoveD2L src));
ins_cost(MEMORY_REF_COST);
size(4);
format %{ "LD $dst, $src \t// MoveD2L" %}
ins_encode( enc_ld(dst, src) );
ins_pipe(pipe_class_memory);
%}
// Move double value from double register to long stack-location.
instruct moveD2L_reg_stack(stackSlotL dst, regD src) %{
match(Set dst (MoveD2L src));
effect(DEF dst, USE src);
ins_cost(MEMORY_REF_COST);
format %{ "STFD $src, $dst \t// MoveD2L" %}
size(4);
ins_encode( enc_stfd(src, dst) );
ins_pipe(pipe_class_memory);
%}
// Move long value from long stack-location to double register.
instruct moveL2D_stack_reg(regD dst, stackSlotL src) %{
match(Set dst (MoveL2D src));
ins_cost(MEMORY_REF_COST);
format %{ "LFD $dst, $src \t// MoveL2D" %}
size(4);
ins_encode( enc_lfd(dst, src) );
ins_pipe(pipe_class_memory);
%}
// Move long value from long register to double stack-location.
instruct moveL2D_reg_stack(stackSlotD dst, iRegLsrc src) %{
match(Set dst (MoveL2D src));
ins_cost(MEMORY_REF_COST);
format %{ "STD $src, $dst \t// MoveL2D" %}
size(4);
ins_encode( enc_std(src, dst) );
ins_pipe(pipe_class_memory);
%}
//----------Register Move Instructions-----------------------------------------
// Replicate for Superword
instruct moveReg(iRegLdst dst, iRegIsrc src) %{
predicate(false);
effect(DEF dst, USE src);
format %{ "MR $dst, $src \t// replicate " %}
// variable size, 0 or 4.
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_or);
__ mr_if_needed($dst$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
//----------Cast instructions (Java-level type cast)---------------------------
// Cast Long to Pointer for unsafe natives.
instruct castX2P(iRegPdst dst, iRegLsrc src) %{
match(Set dst (CastX2P src));
format %{ "MR $dst, $src \t// Long->Ptr" %}
// variable size, 0 or 4.
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_or);
__ mr_if_needed($dst$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Cast Pointer to Long for unsafe natives.
instruct castP2X(iRegLdst dst, iRegP_N2P src) %{
match(Set dst (CastP2X src));
format %{ "MR $dst, $src \t// Ptr->Long" %}
// variable size, 0 or 4.
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_or);
__ mr_if_needed($dst$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct castPP(iRegPdst dst) %{
match(Set dst (CastPP dst));
format %{ " -- \t// castPP of $dst" %}
size(0);
ins_encode( /*empty*/ );
ins_pipe(pipe_class_default);
%}
instruct castII(iRegIdst dst) %{
match(Set dst (CastII dst));
format %{ " -- \t// castII of $dst" %}
size(0);
ins_encode( /*empty*/ );
ins_pipe(pipe_class_default);
%}
instruct checkCastPP(iRegPdst dst) %{
match(Set dst (CheckCastPP dst));
format %{ " -- \t// checkcastPP of $dst" %}
size(0);
ins_encode( /*empty*/ );
ins_pipe(pipe_class_default);
%}
//----------Convert instructions-----------------------------------------------
// Convert to boolean.
// int_to_bool(src) : { 1 if src != 0
// { 0 else
//
// strategy:
// 1) Count leading zeros of 32 bit-value src,
// this returns 32 (0b10.0000) iff src == 0 and <32 otherwise.
// 2) Shift 5 bits to the right, result is 0b1 iff src == 0, 0b0 otherwise.
// 3) Xori the result to get 0b1 if src != 0 and 0b0 if src == 0.
// convI2Bool
instruct convI2Bool_reg__cntlz_Ex(iRegIdst dst, iRegIsrc src) %{
match(Set dst (Conv2B src));
predicate(UseCountLeadingZerosInstructionsPPC64);
ins_cost(DEFAULT_COST);
expand %{
immI shiftAmount %{ 0x5 %}
uimmI16 mask %{ 0x1 %}
iRegIdst tmp1;
iRegIdst tmp2;
countLeadingZerosI(tmp1, src);
urShiftI_reg_imm(tmp2, tmp1, shiftAmount);
xorI_reg_uimm16(dst, tmp2, mask);
%}
%}
instruct convI2Bool_reg__cmove(iRegIdst dst, iRegIsrc src, flagsReg crx) %{
match(Set dst (Conv2B src));
effect(TEMP crx);
predicate(!UseCountLeadingZerosInstructionsPPC64);
ins_cost(DEFAULT_COST);
format %{ "CMPWI $crx, $src, #0 \t// convI2B"
"LI $dst, #0\n\t"
"BEQ $crx, done\n\t"
"LI $dst, #1\n"
"done:" %}
size(16);
ins_encode( enc_convI2B_regI__cmove(dst, src, crx, 0x0, 0x1) );
ins_pipe(pipe_class_compare);
%}
// ConvI2B + XorI
instruct xorI_convI2Bool_reg_immIvalue1__cntlz_Ex(iRegIdst dst, iRegIsrc src, immI_1 mask) %{
match(Set dst (XorI (Conv2B src) mask));
predicate(UseCountLeadingZerosInstructionsPPC64);
ins_cost(DEFAULT_COST);
expand %{
immI shiftAmount %{ 0x5 %}
iRegIdst tmp1;
countLeadingZerosI(tmp1, src);
urShiftI_reg_imm(dst, tmp1, shiftAmount);
%}
%}
instruct xorI_convI2Bool_reg_immIvalue1__cmove(iRegIdst dst, iRegIsrc src, flagsReg crx, immI_1 mask) %{
match(Set dst (XorI (Conv2B src) mask));
effect(TEMP crx);
predicate(!UseCountLeadingZerosInstructionsPPC64);
ins_cost(DEFAULT_COST);
format %{ "CMPWI $crx, $src, #0 \t// Xor(convI2B($src), $mask)"
"LI $dst, #1\n\t"
"BEQ $crx, done\n\t"
"LI $dst, #0\n"
"done:" %}
size(16);
ins_encode( enc_convI2B_regI__cmove(dst, src, crx, 0x1, 0x0) );
ins_pipe(pipe_class_compare);
%}
// AndI 0b0..010..0 + ConvI2B
instruct convI2Bool_andI_reg_immIpowerOf2(iRegIdst dst, iRegIsrc src, immIpowerOf2 mask) %{
match(Set dst (Conv2B (AndI src mask)));
predicate(UseRotateAndMaskInstructionsPPC64);
ins_cost(DEFAULT_COST);
format %{ "RLWINM $dst, $src, $mask \t// convI2B(AndI($src, $mask))" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rlwinm);
__ rlwinm($dst$$Register, $src$$Register, (32-log2_long((jlong)$mask$$constant)) & 0x1f, 31, 31);
%}
ins_pipe(pipe_class_default);
%}
// Convert pointer to boolean.
//
// ptr_to_bool(src) : { 1 if src != 0
// { 0 else
//
// strategy:
// 1) Count leading zeros of 64 bit-value src,
// this returns 64 (0b100.0000) iff src == 0 and <64 otherwise.
// 2) Shift 6 bits to the right, result is 0b1 iff src == 0, 0b0 otherwise.
// 3) Xori the result to get 0b1 if src != 0 and 0b0 if src == 0.
// ConvP2B
instruct convP2Bool_reg__cntlz_Ex(iRegIdst dst, iRegP_N2P src) %{
match(Set dst (Conv2B src));
predicate(UseCountLeadingZerosInstructionsPPC64);
ins_cost(DEFAULT_COST);
expand %{
immI shiftAmount %{ 0x6 %}
uimmI16 mask %{ 0x1 %}
iRegIdst tmp1;
iRegIdst tmp2;
countLeadingZerosP(tmp1, src);
urShiftI_reg_imm(tmp2, tmp1, shiftAmount);
xorI_reg_uimm16(dst, tmp2, mask);
%}
%}
instruct convP2Bool_reg__cmove(iRegIdst dst, iRegP_N2P src, flagsReg crx) %{
match(Set dst (Conv2B src));
effect(TEMP crx);
predicate(!UseCountLeadingZerosInstructionsPPC64);
ins_cost(DEFAULT_COST);
format %{ "CMPDI $crx, $src, #0 \t// convP2B"
"LI $dst, #0\n\t"
"BEQ $crx, done\n\t"
"LI $dst, #1\n"
"done:" %}
size(16);
ins_encode( enc_convP2B_regP__cmove(dst, src, crx, 0x0, 0x1) );
ins_pipe(pipe_class_compare);
%}
// ConvP2B + XorI
instruct xorI_convP2Bool_reg__cntlz_Ex(iRegIdst dst, iRegP_N2P src, immI_1 mask) %{
match(Set dst (XorI (Conv2B src) mask));
predicate(UseCountLeadingZerosInstructionsPPC64);
ins_cost(DEFAULT_COST);
expand %{
immI shiftAmount %{ 0x6 %}
iRegIdst tmp1;
countLeadingZerosP(tmp1, src);
urShiftI_reg_imm(dst, tmp1, shiftAmount);
%}
%}
instruct xorI_convP2Bool_reg_immIvalue1__cmove(iRegIdst dst, iRegP_N2P src, flagsReg crx, immI_1 mask) %{
match(Set dst (XorI (Conv2B src) mask));
effect(TEMP crx);
predicate(!UseCountLeadingZerosInstructionsPPC64);
ins_cost(DEFAULT_COST);
format %{ "CMPDI $crx, $src, #0 \t// XorI(convP2B($src), $mask)"
"LI $dst, #1\n\t"
"BEQ $crx, done\n\t"
"LI $dst, #0\n"
"done:" %}
size(16);
ins_encode( enc_convP2B_regP__cmove(dst, src, crx, 0x1, 0x0) );
ins_pipe(pipe_class_compare);
%}
// if src1 < src2, return -1 else return 0
instruct cmpLTMask_reg_reg_Ex(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
match(Set dst (CmpLTMask src1 src2));
ins_cost(DEFAULT_COST*4);
expand %{
iRegLdst src1s;
iRegLdst src2s;
iRegLdst diff;
convI2L_reg(src1s, src1); // Ensure proper sign extension.
convI2L_reg(src2s, src2); // Ensure proper sign extension.
subL_reg_reg(diff, src1s, src2s);
// Need to consider >=33 bit result, therefore we need signmaskL.
signmask64I_regL(dst, diff);
%}
%}
instruct cmpLTMask_reg_immI0(iRegIdst dst, iRegIsrc src1, immI_0 src2) %{
match(Set dst (CmpLTMask src1 src2)); // if src1 < src2, return -1 else return 0
format %{ "SRAWI $dst, $src1, $src2 \t// CmpLTMask" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_srawi);
__ srawi($dst$$Register, $src1$$Register, 0x1f);
%}
ins_pipe(pipe_class_default);
%}
//----------Arithmetic Conversion Instructions---------------------------------
// Convert to Byte -- nop
// Convert to Short -- nop
// Convert to Int
instruct convB2I_reg(iRegIdst dst, iRegIsrc src, immI_24 amount) %{
match(Set dst (RShiftI (LShiftI src amount) amount));
format %{ "EXTSB $dst, $src \t// byte->int" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_extsb);
__ extsb($dst$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
// LShiftI 16 + RShiftI 16 converts short to int.
instruct convS2I_reg(iRegIdst dst, iRegIsrc src, immI_16 amount) %{
match(Set dst (RShiftI (LShiftI src amount) amount));
format %{ "EXTSH $dst, $src \t// short->int" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_extsh);
__ extsh($dst$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
// ConvL2I + ConvI2L: Sign extend int in long register.
instruct sxtI_L2L_reg(iRegLdst dst, iRegLsrc src) %{
match(Set dst (ConvI2L (ConvL2I src)));
format %{ "EXTSW $dst, $src \t// long->long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_extsw);
__ extsw($dst$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct convL2I_reg(iRegIdst dst, iRegLsrc src) %{
match(Set dst (ConvL2I src));
format %{ "MR $dst, $src \t// long->int" %}
// variable size, 0 or 4
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_or);
__ mr_if_needed($dst$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct convD2IRaw_regD(regD dst, regD src) %{
// no match-rule, false predicate
effect(DEF dst, USE src);
predicate(false);
format %{ "FCTIWZ $dst, $src \t// convD2I, $src != NaN" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fctiwz);;
__ fctiwz($dst$$FloatRegister, $src$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
instruct cmovI_bso_stackSlotL(iRegIdst dst, flagsRegSrc crx, stackSlotL src) %{
// no match-rule, false predicate
effect(DEF dst, USE crx, USE src);
predicate(false);
ins_variable_size_depending_on_alignment(true);
format %{ "cmovI $crx, $dst, $src" %}
// Worst case is branch + move + stop, no stop without scheduler.
size(false /* TODO: PPC PORT(InsertEndGroupPPC64 && Compile::current()->do_hb_scheduling())*/ ? 12 : 8);
ins_encode( enc_cmove_bso_stackSlotL(dst, crx, src) );
ins_pipe(pipe_class_default);
%}
instruct cmovI_bso_stackSlotL_conLvalue0_Ex(iRegIdst dst, flagsRegSrc crx, stackSlotL mem) %{
// no match-rule, false predicate
effect(DEF dst, USE crx, USE mem);
predicate(false);
format %{ "CmovI $dst, $crx, $mem \t// postalloc expanded" %}
postalloc_expand %{
//
// replaces
//
// region dst crx mem
// \ | | /
// dst=cmovI_bso_stackSlotL_conLvalue0
//
// with
//
// region dst
// \ /
// dst=loadConI16(0)
// |
// ^ region dst crx mem
// | \ | | /
// dst=cmovI_bso_stackSlotL
//
// Create new nodes.
MachNode *m1 = new loadConI16Node();
MachNode *m2 = new cmovI_bso_stackSlotLNode();
// inputs for new nodes
m1->add_req(n_region);
m2->add_req(n_region, n_crx, n_mem);
// precedences for new nodes
m2->add_prec(m1);
// operands for new nodes
m1->_opnds[0] = op_dst;
m1->_opnds[1] = new immI16Oper(0);
m2->_opnds[0] = op_dst;
m2->_opnds[1] = op_crx;
m2->_opnds[2] = op_mem;
// registers for new nodes
ra_->set_pair(m1->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this)); // dst
ra_->set_pair(m2->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this)); // dst
// Insert new nodes.
nodes->push(m1);
nodes->push(m2);
%}
%}
// Double to Int conversion, NaN is mapped to 0.
instruct convD2I_reg_ExEx(iRegIdst dst, regD src) %{
match(Set dst (ConvD2I src));
ins_cost(DEFAULT_COST);
expand %{
regD tmpD;
stackSlotL tmpS;
flagsReg crx;
cmpDUnordered_reg_reg(crx, src, src); // Check whether src is NaN.
convD2IRaw_regD(tmpD, src); // Convert float to int (speculated).
moveD2L_reg_stack(tmpS, tmpD); // Store float to stack (speculated).
cmovI_bso_stackSlotL_conLvalue0_Ex(dst, crx, tmpS); // Cmove based on NaN check.
%}
%}
instruct convF2IRaw_regF(regF dst, regF src) %{
// no match-rule, false predicate
effect(DEF dst, USE src);
predicate(false);
format %{ "FCTIWZ $dst, $src \t// convF2I, $src != NaN" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fctiwz);
__ fctiwz($dst$$FloatRegister, $src$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
// Float to Int conversion, NaN is mapped to 0.
instruct convF2I_regF_ExEx(iRegIdst dst, regF src) %{
match(Set dst (ConvF2I src));
ins_cost(DEFAULT_COST);
expand %{
regF tmpF;
stackSlotL tmpS;
flagsReg crx;
cmpFUnordered_reg_reg(crx, src, src); // Check whether src is NaN.
convF2IRaw_regF(tmpF, src); // Convert float to int (speculated).
moveF2L_reg_stack(tmpS, tmpF); // Store float to stack (speculated).
cmovI_bso_stackSlotL_conLvalue0_Ex(dst, crx, tmpS); // Cmove based on NaN check.
%}
%}
// Convert to Long
instruct convI2L_reg(iRegLdst dst, iRegIsrc src) %{
match(Set dst (ConvI2L src));
format %{ "EXTSW $dst, $src \t// int->long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_extsw);
__ extsw($dst$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Zero-extend: convert unsigned int to long (convUI2L).
instruct zeroExtendL_regI(iRegLdst dst, iRegIsrc src, immL_32bits mask) %{
match(Set dst (AndL (ConvI2L src) mask));
ins_cost(DEFAULT_COST);
format %{ "CLRLDI $dst, $src, #32 \t// zero-extend int to long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicl);
__ clrldi($dst$$Register, $src$$Register, 32);
%}
ins_pipe(pipe_class_default);
%}
// Zero-extend: convert unsigned int to long in long register.
instruct zeroExtendL_regL(iRegLdst dst, iRegLsrc src, immL_32bits mask) %{
match(Set dst (AndL src mask));
ins_cost(DEFAULT_COST);
format %{ "CLRLDI $dst, $src, #32 \t// zero-extend int to long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicl);
__ clrldi($dst$$Register, $src$$Register, 32);
%}
ins_pipe(pipe_class_default);
%}
instruct convF2LRaw_regF(regF dst, regF src) %{
// no match-rule, false predicate
effect(DEF dst, USE src);
predicate(false);
format %{ "FCTIDZ $dst, $src \t// convF2L, $src != NaN" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fctiwz);
__ fctidz($dst$$FloatRegister, $src$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
instruct cmovL_bso_stackSlotL(iRegLdst dst, flagsRegSrc crx, stackSlotL src) %{
// no match-rule, false predicate
effect(DEF dst, USE crx, USE src);
predicate(false);
ins_variable_size_depending_on_alignment(true);
format %{ "cmovL $crx, $dst, $src" %}
// Worst case is branch + move + stop, no stop without scheduler.
size(false /* TODO: PPC PORT Compile::current()->do_hb_scheduling()*/ ? 12 : 8);
ins_encode( enc_cmove_bso_stackSlotL(dst, crx, src) );
ins_pipe(pipe_class_default);
%}
instruct cmovL_bso_stackSlotL_conLvalue0_Ex(iRegLdst dst, flagsRegSrc crx, stackSlotL mem) %{
// no match-rule, false predicate
effect(DEF dst, USE crx, USE mem);
predicate(false);
format %{ "CmovL $dst, $crx, $mem \t// postalloc expanded" %}
postalloc_expand %{
//
// replaces
//
// region dst crx mem
// \ | | /
// dst=cmovL_bso_stackSlotL_conLvalue0
//
// with
//
// region dst
// \ /
// dst=loadConL16(0)
// |
// ^ region dst crx mem
// | \ | | /
// dst=cmovL_bso_stackSlotL
//
// Create new nodes.
MachNode *m1 = new loadConL16Node();
MachNode *m2 = new cmovL_bso_stackSlotLNode();
// inputs for new nodes
m1->add_req(n_region);
m2->add_req(n_region, n_crx, n_mem);
m2->add_prec(m1);
// operands for new nodes
m1->_opnds[0] = op_dst;
m1->_opnds[1] = new immL16Oper(0);
m2->_opnds[0] = op_dst;
m2->_opnds[1] = op_crx;
m2->_opnds[2] = op_mem;
// registers for new nodes
ra_->set_pair(m1->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this)); // dst
ra_->set_pair(m2->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this)); // dst
// Insert new nodes.
nodes->push(m1);
nodes->push(m2);
%}
%}
// Float to Long conversion, NaN is mapped to 0.
instruct convF2L_reg_ExEx(iRegLdst dst, regF src) %{
match(Set dst (ConvF2L src));
ins_cost(DEFAULT_COST);
expand %{
regF tmpF;
stackSlotL tmpS;
flagsReg crx;
cmpFUnordered_reg_reg(crx, src, src); // Check whether src is NaN.
convF2LRaw_regF(tmpF, src); // Convert float to long (speculated).
moveF2L_reg_stack(tmpS, tmpF); // Store float to stack (speculated).
cmovL_bso_stackSlotL_conLvalue0_Ex(dst, crx, tmpS); // Cmove based on NaN check.
%}
%}
instruct convD2LRaw_regD(regD dst, regD src) %{
// no match-rule, false predicate
effect(DEF dst, USE src);
predicate(false);
format %{ "FCTIDZ $dst, $src \t// convD2L $src != NaN" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fctiwz);
__ fctidz($dst$$FloatRegister, $src$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
// Double to Long conversion, NaN is mapped to 0.
instruct convD2L_reg_ExEx(iRegLdst dst, regD src) %{
match(Set dst (ConvD2L src));
ins_cost(DEFAULT_COST);
expand %{
regD tmpD;
stackSlotL tmpS;
flagsReg crx;
cmpDUnordered_reg_reg(crx, src, src); // Check whether src is NaN.
convD2LRaw_regD(tmpD, src); // Convert float to long (speculated).
moveD2L_reg_stack(tmpS, tmpD); // Store float to stack (speculated).
cmovL_bso_stackSlotL_conLvalue0_Ex(dst, crx, tmpS); // Cmove based on NaN check.
%}
%}
// Convert to Float
// Placed here as needed in expand.
instruct convL2DRaw_regD(regD dst, regD src) %{
// no match-rule, false predicate
effect(DEF dst, USE src);
predicate(false);
format %{ "FCFID $dst, $src \t// convL2D" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fcfid);
__ fcfid($dst$$FloatRegister, $src$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
// Placed here as needed in expand.
instruct convD2F_reg(regF dst, regD src) %{
match(Set dst (ConvD2F src));
format %{ "FRSP $dst, $src \t// convD2F" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_frsp);
__ frsp($dst$$FloatRegister, $src$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
// Integer to Float conversion.
instruct convI2F_ireg_Ex(regF dst, iRegIsrc src) %{
match(Set dst (ConvI2F src));
predicate(!VM_Version::has_fcfids());
ins_cost(DEFAULT_COST);
expand %{
iRegLdst tmpL;
stackSlotL tmpS;
regD tmpD;
regD tmpD2;
convI2L_reg(tmpL, src); // Sign-extension int to long.
regL_to_stkL(tmpS, tmpL); // Store long to stack.
moveL2D_stack_reg(tmpD, tmpS); // Load long into double register.
convL2DRaw_regD(tmpD2, tmpD); // Convert to double.
convD2F_reg(dst, tmpD2); // Convert double to float.
%}
%}
instruct convL2FRaw_regF(regF dst, regD src) %{
// no match-rule, false predicate
effect(DEF dst, USE src);
predicate(false);
format %{ "FCFIDS $dst, $src \t// convL2F" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fcfid);
__ fcfids($dst$$FloatRegister, $src$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
// Integer to Float conversion. Special version for Power7.
instruct convI2F_ireg_fcfids_Ex(regF dst, iRegIsrc src) %{
match(Set dst (ConvI2F src));
predicate(VM_Version::has_fcfids());
ins_cost(DEFAULT_COST);
expand %{
iRegLdst tmpL;
stackSlotL tmpS;
regD tmpD;
convI2L_reg(tmpL, src); // Sign-extension int to long.
regL_to_stkL(tmpS, tmpL); // Store long to stack.
moveL2D_stack_reg(tmpD, tmpS); // Load long into double register.
convL2FRaw_regF(dst, tmpD); // Convert to float.
%}
%}
// L2F to avoid runtime call.
instruct convL2F_ireg_fcfids_Ex(regF dst, iRegLsrc src) %{
match(Set dst (ConvL2F src));
predicate(VM_Version::has_fcfids());
ins_cost(DEFAULT_COST);
expand %{
stackSlotL tmpS;
regD tmpD;
regL_to_stkL(tmpS, src); // Store long to stack.
moveL2D_stack_reg(tmpD, tmpS); // Load long into double register.
convL2FRaw_regF(dst, tmpD); // Convert to float.
%}
%}
// Moved up as used in expand.
//instruct convD2F_reg(regF dst, regD src) %{%}
// Convert to Double
// Integer to Double conversion.
instruct convI2D_reg_Ex(regD dst, iRegIsrc src) %{
match(Set dst (ConvI2D src));
ins_cost(DEFAULT_COST);
expand %{
iRegLdst tmpL;
stackSlotL tmpS;
regD tmpD;
convI2L_reg(tmpL, src); // Sign-extension int to long.
regL_to_stkL(tmpS, tmpL); // Store long to stack.
moveL2D_stack_reg(tmpD, tmpS); // Load long into double register.
convL2DRaw_regD(dst, tmpD); // Convert to double.
%}
%}
// Long to Double conversion
instruct convL2D_reg_Ex(regD dst, stackSlotL src) %{
match(Set dst (ConvL2D src));
ins_cost(DEFAULT_COST + MEMORY_REF_COST);
expand %{
regD tmpD;
moveL2D_stack_reg(tmpD, src);
convL2DRaw_regD(dst, tmpD);
%}
%}
instruct convF2D_reg(regD dst, regF src) %{
match(Set dst (ConvF2D src));
format %{ "FMR $dst, $src \t// float->double" %}
// variable size, 0 or 4
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fmr);
__ fmr_if_needed($dst$$FloatRegister, $src$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
//----------Control Flow Instructions------------------------------------------
// Compare Instructions
// Compare Integers
instruct cmpI_reg_reg(flagsReg crx, iRegIsrc src1, iRegIsrc src2) %{
match(Set crx (CmpI src1 src2));
size(4);
format %{ "CMPW $crx, $src1, $src2" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_cmp);
__ cmpw($crx$$CondRegister, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_compare);
%}
instruct cmpI_reg_imm16(flagsReg crx, iRegIsrc src1, immI16 src2) %{
match(Set crx (CmpI src1 src2));
format %{ "CMPWI $crx, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_cmpi);
__ cmpwi($crx$$CondRegister, $src1$$Register, $src2$$constant);
%}
ins_pipe(pipe_class_compare);
%}
// (src1 & src2) == 0?
instruct testI_reg_imm(flagsRegCR0 cr0, iRegIsrc src1, uimmI16 src2, immI_0 zero) %{
match(Set cr0 (CmpI (AndI src1 src2) zero));
// r0 is killed
format %{ "ANDI R0, $src1, $src2 \t// BTST int" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_andi_);
__ andi_(R0, $src1$$Register, $src2$$constant);
%}
ins_pipe(pipe_class_compare);
%}
instruct cmpL_reg_reg(flagsReg crx, iRegLsrc src1, iRegLsrc src2) %{
match(Set crx (CmpL src1 src2));
format %{ "CMPD $crx, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_cmp);
__ cmpd($crx$$CondRegister, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_compare);
%}
instruct cmpL_reg_imm16(flagsReg crx, iRegLsrc src1, immL16 src2) %{
match(Set crx (CmpL src1 src2));
format %{ "CMPDI $crx, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_cmpi);
__ cmpdi($crx$$CondRegister, $src1$$Register, $src2$$constant);
%}
ins_pipe(pipe_class_compare);
%}
instruct testL_reg_reg(flagsRegCR0 cr0, iRegLsrc src1, iRegLsrc src2, immL_0 zero) %{
match(Set cr0 (CmpL (AndL src1 src2) zero));
// r0 is killed
format %{ "AND R0, $src1, $src2 \t// BTST long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_and_);
__ and_(R0, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_compare);
%}
instruct testL_reg_imm(flagsRegCR0 cr0, iRegLsrc src1, uimmL16 src2, immL_0 zero) %{
match(Set cr0 (CmpL (AndL src1 src2) zero));
// r0 is killed
format %{ "ANDI R0, $src1, $src2 \t// BTST long" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_andi_);
__ andi_(R0, $src1$$Register, $src2$$constant);
%}
ins_pipe(pipe_class_compare);
%}
instruct cmovI_conIvalueMinus1_conIvalue1(iRegIdst dst, flagsRegSrc crx) %{
// no match-rule, false predicate
effect(DEF dst, USE crx);
predicate(false);
ins_variable_size_depending_on_alignment(true);
format %{ "cmovI $crx, $dst, -1, 0, +1" %}
// Worst case is branch + move + branch + move + stop, no stop without scheduler.
size(false /* TODO: PPC PORTInsertEndGroupPPC64 && Compile::current()->do_hb_scheduling())*/ ? 20 : 16);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_cmove);
Label done;
// li(Rdst, 0); // equal -> 0
__ beq($crx$$CondRegister, done);
__ li($dst$$Register, 1); // greater -> +1
__ bgt($crx$$CondRegister, done);
__ li($dst$$Register, -1); // unordered or less -> -1
// TODO: PPC port__ endgroup_if_needed(_size == 20);
__ bind(done);
%}
ins_pipe(pipe_class_compare);
%}
instruct cmovI_conIvalueMinus1_conIvalue0_conIvalue1_Ex(iRegIdst dst, flagsRegSrc crx) %{
// no match-rule, false predicate
effect(DEF dst, USE crx);
predicate(false);
format %{ "CmovI $crx, $dst, -1, 0, +1 \t// postalloc expanded" %}
postalloc_expand %{
//
// replaces
//
// region crx
// \ |
// dst=cmovI_conIvalueMinus1_conIvalue0_conIvalue1
//
// with
//
// region
// \
// dst=loadConI16(0)
// |
// ^ region crx
// | \ |
// dst=cmovI_conIvalueMinus1_conIvalue1
//
// Create new nodes.
MachNode *m1 = new loadConI16Node();
MachNode *m2 = new cmovI_conIvalueMinus1_conIvalue1Node();
// inputs for new nodes
m1->add_req(n_region);
m2->add_req(n_region, n_crx);
m2->add_prec(m1);
// operands for new nodes
m1->_opnds[0] = op_dst;
m1->_opnds[1] = new immI16Oper(0);
m2->_opnds[0] = op_dst;
m2->_opnds[1] = op_crx;
// registers for new nodes
ra_->set_pair(m1->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this)); // dst
ra_->set_pair(m2->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this)); // dst
// Insert new nodes.
nodes->push(m1);
nodes->push(m2);
%}
%}
// Manifest a CmpL3 result in an integer register. Very painful.
// This is the test to avoid.
// (src1 < src2) ? -1 : ((src1 > src2) ? 1 : 0)
instruct cmpL3_reg_reg_ExEx(iRegIdst dst, iRegLsrc src1, iRegLsrc src2) %{
match(Set dst (CmpL3 src1 src2));
ins_cost(DEFAULT_COST*5+BRANCH_COST);
expand %{
flagsReg tmp1;
cmpL_reg_reg(tmp1, src1, src2);
cmovI_conIvalueMinus1_conIvalue0_conIvalue1_Ex(dst, tmp1);
%}
%}
// Implicit range checks.
// A range check in the ideal world has one of the following shapes:
// - (If le (CmpU length index)), (IfTrue throw exception)
// - (If lt (CmpU index length)), (IfFalse throw exception)
//
// Match range check 'If le (CmpU length index)'.
instruct rangeCheck_iReg_uimm15(cmpOp cmp, iRegIsrc src_length, uimmI15 index, label labl) %{
match(If cmp (CmpU src_length index));
effect(USE labl);
predicate(TrapBasedRangeChecks &&
_kids[0]->_leaf->as_Bool()->_test._test == BoolTest::le &&
PROB_UNLIKELY(_leaf->as_If()->_prob) >= PROB_ALWAYS &&
(Matcher::branches_to_uncommon_trap(_leaf)));
ins_is_TrapBasedCheckNode(true);
format %{ "TWI $index $cmp $src_length \t// RangeCheck => trap $labl" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_twi);
if ($cmp$$cmpcode == 0x1 /* less_equal */) {
__ trap_range_check_le($src_length$$Register, $index$$constant);
} else {
// Both successors are uncommon traps, probability is 0.
// Node got flipped during fixup flow.
assert($cmp$$cmpcode == 0x9, "must be greater");
__ trap_range_check_g($src_length$$Register, $index$$constant);
}
%}
ins_pipe(pipe_class_trap);
%}
// Match range check 'If lt (CmpU index length)'.
instruct rangeCheck_iReg_iReg(cmpOp cmp, iRegIsrc src_index, iRegIsrc src_length, label labl) %{
match(If cmp (CmpU src_index src_length));
effect(USE labl);
predicate(TrapBasedRangeChecks &&
_kids[0]->_leaf->as_Bool()->_test._test == BoolTest::lt &&
_leaf->as_If()->_prob >= PROB_ALWAYS &&
(Matcher::branches_to_uncommon_trap(_leaf)));
ins_is_TrapBasedCheckNode(true);
format %{ "TW $src_index $cmp $src_length \t// RangeCheck => trap $labl" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_tw);
if ($cmp$$cmpcode == 0x0 /* greater_equal */) {
__ trap_range_check_ge($src_index$$Register, $src_length$$Register);
} else {
// Both successors are uncommon traps, probability is 0.
// Node got flipped during fixup flow.
assert($cmp$$cmpcode == 0x8, "must be less");
__ trap_range_check_l($src_index$$Register, $src_length$$Register);
}
%}
ins_pipe(pipe_class_trap);
%}
// Match range check 'If lt (CmpU index length)'.
instruct rangeCheck_uimm15_iReg(cmpOp cmp, iRegIsrc src_index, uimmI15 length, label labl) %{
match(If cmp (CmpU src_index length));
effect(USE labl);
predicate(TrapBasedRangeChecks &&
_kids[0]->_leaf->as_Bool()->_test._test == BoolTest::lt &&
_leaf->as_If()->_prob >= PROB_ALWAYS &&
(Matcher::branches_to_uncommon_trap(_leaf)));
ins_is_TrapBasedCheckNode(true);
format %{ "TWI $src_index $cmp $length \t// RangeCheck => trap $labl" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_twi);
if ($cmp$$cmpcode == 0x0 /* greater_equal */) {
__ trap_range_check_ge($src_index$$Register, $length$$constant);
} else {
// Both successors are uncommon traps, probability is 0.
// Node got flipped during fixup flow.
assert($cmp$$cmpcode == 0x8, "must be less");
__ trap_range_check_l($src_index$$Register, $length$$constant);
}
%}
ins_pipe(pipe_class_trap);
%}
instruct compU_reg_reg(flagsReg crx, iRegIsrc src1, iRegIsrc src2) %{
match(Set crx (CmpU src1 src2));
format %{ "CMPLW $crx, $src1, $src2 \t// unsigned" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_cmpl);
__ cmplw($crx$$CondRegister, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_compare);
%}
instruct compU_reg_uimm16(flagsReg crx, iRegIsrc src1, uimmI16 src2) %{
match(Set crx (CmpU src1 src2));
size(4);
format %{ "CMPLWI $crx, $src1, $src2" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_cmpli);
__ cmplwi($crx$$CondRegister, $src1$$Register, $src2$$constant);
%}
ins_pipe(pipe_class_compare);
%}
// Implicit zero checks (more implicit null checks).
// No constant pool entries required.
instruct zeroCheckN_iReg_imm0(cmpOp cmp, iRegNsrc value, immN_0 zero, label labl) %{
match(If cmp (CmpN value zero));
effect(USE labl);
predicate(TrapBasedNullChecks &&
_kids[0]->_leaf->as_Bool()->_test._test == BoolTest::ne &&
_leaf->as_If()->_prob >= PROB_LIKELY_MAG(4) &&
Matcher::branches_to_uncommon_trap(_leaf));
ins_cost(1);
ins_is_TrapBasedCheckNode(true);
format %{ "TDI $value $cmp $zero \t// ZeroCheckN => trap $labl" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_tdi);
if ($cmp$$cmpcode == 0xA) {
__ trap_null_check($value$$Register);
} else {
// Both successors are uncommon traps, probability is 0.
// Node got flipped during fixup flow.
assert($cmp$$cmpcode == 0x2 , "must be equal(0xA) or notEqual(0x2)");
__ trap_null_check($value$$Register, Assembler::traptoGreaterThanUnsigned);
}
%}
ins_pipe(pipe_class_trap);
%}
// Compare narrow oops.
instruct cmpN_reg_reg(flagsReg crx, iRegNsrc src1, iRegNsrc src2) %{
match(Set crx (CmpN src1 src2));
size(4);
ins_cost(2);
format %{ "CMPLW $crx, $src1, $src2 \t// compressed ptr" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_cmpl);
__ cmplw($crx$$CondRegister, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_compare);
%}
instruct cmpN_reg_imm0(flagsReg crx, iRegNsrc src1, immN_0 src2) %{
match(Set crx (CmpN src1 src2));
// Make this more expensive than zeroCheckN_iReg_imm0.
ins_cost(2);
format %{ "CMPLWI $crx, $src1, $src2 \t// compressed ptr" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_cmpli);
__ cmplwi($crx$$CondRegister, $src1$$Register, $src2$$constant);
%}
ins_pipe(pipe_class_compare);
%}
// Implicit zero checks (more implicit null checks).
// No constant pool entries required.
instruct zeroCheckP_reg_imm0(cmpOp cmp, iRegP_N2P value, immP_0 zero, label labl) %{
match(If cmp (CmpP value zero));
effect(USE labl);
predicate(TrapBasedNullChecks &&
_kids[0]->_leaf->as_Bool()->_test._test == BoolTest::ne &&
_leaf->as_If()->_prob >= PROB_LIKELY_MAG(4) &&
Matcher::branches_to_uncommon_trap(_leaf));
ins_cost(1); // Should not be cheaper than zeroCheckN.
ins_is_TrapBasedCheckNode(true);
format %{ "TDI $value $cmp $zero \t// ZeroCheckP => trap $labl" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_tdi);
if ($cmp$$cmpcode == 0xA) {
__ trap_null_check($value$$Register);
} else {
// Both successors are uncommon traps, probability is 0.
// Node got flipped during fixup flow.
assert($cmp$$cmpcode == 0x2 , "must be equal(0xA) or notEqual(0x2)");
__ trap_null_check($value$$Register, Assembler::traptoGreaterThanUnsigned);
}
%}
ins_pipe(pipe_class_trap);
%}
// Compare Pointers
instruct cmpP_reg_reg(flagsReg crx, iRegP_N2P src1, iRegP_N2P src2) %{
match(Set crx (CmpP src1 src2));
format %{ "CMPLD $crx, $src1, $src2 \t// ptr" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_cmpl);
__ cmpld($crx$$CondRegister, $src1$$Register, $src2$$Register);
%}
ins_pipe(pipe_class_compare);
%}
// Used in postalloc expand.
instruct cmpP_reg_imm16(flagsReg crx, iRegPsrc src1, immL16 src2) %{
// This match rule prevents reordering of node before a safepoint.
// This only makes sense if this instructions is used exclusively
// for the expansion of EncodeP!
match(Set crx (CmpP src1 src2));
predicate(false);
format %{ "CMPDI $crx, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_cmpi);
__ cmpdi($crx$$CondRegister, $src1$$Register, $src2$$constant);
%}
ins_pipe(pipe_class_compare);
%}
//----------Float Compares----------------------------------------------------
instruct cmpFUnordered_reg_reg(flagsReg crx, regF src1, regF src2) %{
// Needs matchrule, see cmpDUnordered.
match(Set crx (CmpF src1 src2));
// no match-rule, false predicate
predicate(false);
format %{ "cmpFUrd $crx, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fcmpu);
__ fcmpu($crx$$CondRegister, $src1$$FloatRegister, $src2$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
instruct cmov_bns_less(flagsReg crx) %{
// no match-rule, false predicate
effect(DEF crx);
predicate(false);
ins_variable_size_depending_on_alignment(true);
format %{ "cmov $crx" %}
// Worst case is branch + move + stop, no stop without scheduler.
size(false /* TODO: PPC PORT(InsertEndGroupPPC64 && Compile::current()->do_hb_scheduling())*/ ? 16 : 12);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_cmovecr);
Label done;
__ bns($crx$$CondRegister, done); // not unordered -> keep crx
__ li(R0, 0);
__ cmpwi($crx$$CondRegister, R0, 1); // unordered -> set crx to 'less'
// TODO PPC port __ endgroup_if_needed(_size == 16);
__ bind(done);
%}
ins_pipe(pipe_class_default);
%}
// Compare floating, generate condition code.
instruct cmpF_reg_reg_Ex(flagsReg crx, regF src1, regF src2) %{
// FIXME: should we match 'If cmp (CmpF src1 src2))' ??
//
// The following code sequence occurs a lot in mpegaudio:
//
// block BXX:
// 0: instruct cmpFUnordered_reg_reg (cmpF_reg_reg-0):
// cmpFUrd CCR6, F11, F9
// 4: instruct cmov_bns_less (cmpF_reg_reg-1):
// cmov CCR6
// 8: instruct branchConSched:
// B_FARle CCR6, B56 P=0.500000 C=-1.000000
match(Set crx (CmpF src1 src2));
ins_cost(DEFAULT_COST+BRANCH_COST);
format %{ "CmpF $crx, $src1, $src2 \t// postalloc expanded" %}
postalloc_expand %{
//
// replaces
//
// region src1 src2
// \ | |
// crx=cmpF_reg_reg
//
// with
//
// region src1 src2
// \ | |
// crx=cmpFUnordered_reg_reg
// |
// ^ region
// | \
// crx=cmov_bns_less
//
// Create new nodes.
MachNode *m1 = new cmpFUnordered_reg_regNode();
MachNode *m2 = new cmov_bns_lessNode();
// inputs for new nodes
m1->add_req(n_region, n_src1, n_src2);
m2->add_req(n_region);
m2->add_prec(m1);
// operands for new nodes
m1->_opnds[0] = op_crx;
m1->_opnds[1] = op_src1;
m1->_opnds[2] = op_src2;
m2->_opnds[0] = op_crx;
// registers for new nodes
ra_->set_pair(m1->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this)); // crx
ra_->set_pair(m2->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this)); // crx
// Insert new nodes.
nodes->push(m1);
nodes->push(m2);
%}
%}
// Compare float, generate -1,0,1
instruct cmpF3_reg_reg_ExEx(iRegIdst dst, regF src1, regF src2) %{
match(Set dst (CmpF3 src1 src2));
ins_cost(DEFAULT_COST*5+BRANCH_COST);
expand %{
flagsReg tmp1;
cmpFUnordered_reg_reg(tmp1, src1, src2);
cmovI_conIvalueMinus1_conIvalue0_conIvalue1_Ex(dst, tmp1);
%}
%}
instruct cmpDUnordered_reg_reg(flagsReg crx, regD src1, regD src2) %{
// Needs matchrule so that ideal opcode is Cmp. This causes that gcm places the
// node right before the conditional move using it.
// In jck test api/java_awt/geom/QuadCurve2DFloat/index.html#SetCurveTesttestCase7,
// compilation of java.awt.geom.RectangularShape::getBounds()Ljava/awt/Rectangle
// crashed in register allocation where the flags Reg between cmpDUnoredered and a
// conditional move was supposed to be spilled.
match(Set crx (CmpD src1 src2));
// False predicate, shall not be matched.
predicate(false);
format %{ "cmpFUrd $crx, $src1, $src2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fcmpu);
__ fcmpu($crx$$CondRegister, $src1$$FloatRegister, $src2$$FloatRegister);
%}
ins_pipe(pipe_class_default);
%}
instruct cmpD_reg_reg_Ex(flagsReg crx, regD src1, regD src2) %{
match(Set crx (CmpD src1 src2));
ins_cost(DEFAULT_COST+BRANCH_COST);
format %{ "CmpD $crx, $src1, $src2 \t// postalloc expanded" %}
postalloc_expand %{
//
// replaces
//
// region src1 src2
// \ | |
// crx=cmpD_reg_reg
//
// with
//
// region src1 src2
// \ | |
// crx=cmpDUnordered_reg_reg
// |
// ^ region
// | \
// crx=cmov_bns_less
//
// create new nodes
MachNode *m1 = new cmpDUnordered_reg_regNode();
MachNode *m2 = new cmov_bns_lessNode();
// inputs for new nodes
m1->add_req(n_region, n_src1, n_src2);
m2->add_req(n_region);
m2->add_prec(m1);
// operands for new nodes
m1->_opnds[0] = op_crx;
m1->_opnds[1] = op_src1;
m1->_opnds[2] = op_src2;
m2->_opnds[0] = op_crx;
// registers for new nodes
ra_->set_pair(m1->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this)); // crx
ra_->set_pair(m2->_idx, ra_->get_reg_second(this), ra_->get_reg_first(this)); // crx
// Insert new nodes.
nodes->push(m1);
nodes->push(m2);
%}
%}
// Compare double, generate -1,0,1
instruct cmpD3_reg_reg_ExEx(iRegIdst dst, regD src1, regD src2) %{
match(Set dst (CmpD3 src1 src2));
ins_cost(DEFAULT_COST*5+BRANCH_COST);
expand %{
flagsReg tmp1;
cmpDUnordered_reg_reg(tmp1, src1, src2);
cmovI_conIvalueMinus1_conIvalue0_conIvalue1_Ex(dst, tmp1);
%}
%}
//----------Branches---------------------------------------------------------
// Jump
// Direct Branch.
instruct branch(label labl) %{
match(Goto);
effect(USE labl);
ins_cost(BRANCH_COST);
format %{ "B $labl" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_b);
Label d; // dummy
__ bind(d);
Label* p = $labl$$label;
// `p' is `NULL' when this encoding class is used only to
// determine the size of the encoded instruction.
Label& l = (NULL == p)? d : *(p);
__ b(l);
%}
ins_pipe(pipe_class_default);
%}
// Conditional Near Branch
instruct branchCon(cmpOp cmp, flagsRegSrc crx, label lbl) %{
// Same match rule as `branchConFar'.
match(If cmp crx);
effect(USE lbl);
ins_cost(BRANCH_COST);
// If set to 1 this indicates that the current instruction is a
// short variant of a long branch. This avoids using this
// instruction in first-pass matching. It will then only be used in
// the `Shorten_branches' pass.
ins_short_branch(1);
format %{ "B$cmp $crx, $lbl" %}
size(4);
ins_encode( enc_bc(crx, cmp, lbl) );
ins_pipe(pipe_class_default);
%}
// This is for cases when the ppc64 `bc' instruction does not
// reach far enough. So we emit a far branch here, which is more
// expensive.
//
// Conditional Far Branch
instruct branchConFar(cmpOp cmp, flagsRegSrc crx, label lbl) %{
// Same match rule as `branchCon'.
match(If cmp crx);
effect(USE crx, USE lbl);
predicate(!false /* TODO: PPC port HB_Schedule*/);
// Higher cost than `branchCon'.
ins_cost(5*BRANCH_COST);
// This is not a short variant of a branch, but the long variant.
ins_short_branch(0);
format %{ "B_FAR$cmp $crx, $lbl" %}
size(8);
ins_encode( enc_bc_far(crx, cmp, lbl) );
ins_pipe(pipe_class_default);
%}
// Conditional Branch used with Power6 scheduler (can be far or short).
instruct branchConSched(cmpOp cmp, flagsRegSrc crx, label lbl) %{
// Same match rule as `branchCon'.
match(If cmp crx);
effect(USE crx, USE lbl);
predicate(false /* TODO: PPC port HB_Schedule*/);
// Higher cost than `branchCon'.
ins_cost(5*BRANCH_COST);
// Actually size doesn't depend on alignment but on shortening.
ins_variable_size_depending_on_alignment(true);
// long variant.
ins_short_branch(0);
format %{ "B_FAR$cmp $crx, $lbl" %}
size(8); // worst case
ins_encode( enc_bc_short_far(crx, cmp, lbl) );
ins_pipe(pipe_class_default);
%}
instruct branchLoopEnd(cmpOp cmp, flagsRegSrc crx, label labl) %{
match(CountedLoopEnd cmp crx);
effect(USE labl);
ins_cost(BRANCH_COST);
// short variant.
ins_short_branch(1);
format %{ "B$cmp $crx, $labl \t// counted loop end" %}
size(4);
ins_encode( enc_bc(crx, cmp, labl) );
ins_pipe(pipe_class_default);
%}
instruct branchLoopEndFar(cmpOp cmp, flagsRegSrc crx, label labl) %{
match(CountedLoopEnd cmp crx);
effect(USE labl);
predicate(!false /* TODO: PPC port HB_Schedule */);
ins_cost(BRANCH_COST);
// Long variant.
ins_short_branch(0);
format %{ "B_FAR$cmp $crx, $labl \t// counted loop end" %}
size(8);
ins_encode( enc_bc_far(crx, cmp, labl) );
ins_pipe(pipe_class_default);
%}
// Conditional Branch used with Power6 scheduler (can be far or short).
instruct branchLoopEndSched(cmpOp cmp, flagsRegSrc crx, label labl) %{
match(CountedLoopEnd cmp crx);
effect(USE labl);
predicate(false /* TODO: PPC port HB_Schedule */);
// Higher cost than `branchCon'.
ins_cost(5*BRANCH_COST);
// Actually size doesn't depend on alignment but on shortening.
ins_variable_size_depending_on_alignment(true);
// Long variant.
ins_short_branch(0);
format %{ "B_FAR$cmp $crx, $labl \t// counted loop end" %}
size(8); // worst case
ins_encode( enc_bc_short_far(crx, cmp, labl) );
ins_pipe(pipe_class_default);
%}
// ============================================================================
// Java runtime operations, intrinsics and other complex operations.
// The 2nd slow-half of a subtype check. Scan the subklass's 2ndary superklass
// array for an instance of the superklass. Set a hidden internal cache on a
// hit (cache is checked with exposed code in gen_subtype_check()). Return
// not zero for a miss or zero for a hit. The encoding ALSO sets flags.
//
// GL TODO: Improve this.
// - result should not be a TEMP
// - Add match rule as on sparc avoiding additional Cmp.
instruct partialSubtypeCheck(iRegPdst result, iRegP_N2P subklass, iRegP_N2P superklass,
iRegPdst tmp_klass, iRegPdst tmp_arrayptr) %{
match(Set result (PartialSubtypeCheck subklass superklass));
effect(TEMP_DEF result, TEMP tmp_klass, TEMP tmp_arrayptr);
ins_cost(DEFAULT_COST*10);
format %{ "PartialSubtypeCheck $result = ($subklass instanceOf $superklass) tmp: $tmp_klass, $tmp_arrayptr" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ check_klass_subtype_slow_path($subklass$$Register, $superklass$$Register, $tmp_arrayptr$$Register,
$tmp_klass$$Register, NULL, $result$$Register);
%}
ins_pipe(pipe_class_default);
%}
// inlined locking and unlocking
instruct cmpFastLock(flagsReg crx, iRegPdst oop, iRegPdst box, iRegPdst tmp1, iRegPdst tmp2) %{
match(Set crx (FastLock oop box));
effect(TEMP tmp1, TEMP tmp2);
predicate(!Compile::current()->use_rtm());
format %{ "FASTLOCK $oop, $box, $tmp1, $tmp2" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ compiler_fast_lock_object($crx$$CondRegister, $oop$$Register, $box$$Register,
$tmp1$$Register, $tmp2$$Register, /*tmp3*/ R0,
UseBiasedLocking && !UseOptoBiasInlining);
// If locking was successfull, crx should indicate 'EQ'.
// The compiler generates a branch to the runtime call to
// _complete_monitor_locking_Java for the case where crx is 'NE'.
%}
ins_pipe(pipe_class_compare);
%}
// Separate version for TM. Use bound register for box to enable USE_KILL.
instruct cmpFastLock_tm(flagsReg crx, iRegPdst oop, rarg2RegP box, iRegPdst tmp1, iRegPdst tmp2, iRegPdst tmp3) %{
match(Set crx (FastLock oop box));
effect(TEMP tmp1, TEMP tmp2, TEMP tmp3, USE_KILL box);
predicate(Compile::current()->use_rtm());
format %{ "FASTLOCK $oop, $box, $tmp1, $tmp2, $tmp3 (TM)" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ compiler_fast_lock_object($crx$$CondRegister, $oop$$Register, $box$$Register,
$tmp1$$Register, $tmp2$$Register, $tmp3$$Register,
/*Biased Locking*/ false,
_rtm_counters, _stack_rtm_counters,
((Method*)(ra_->C->method()->constant_encoding()))->method_data(),
/*TM*/ true, ra_->C->profile_rtm());
// If locking was successfull, crx should indicate 'EQ'.
// The compiler generates a branch to the runtime call to
// _complete_monitor_locking_Java for the case where crx is 'NE'.
%}
ins_pipe(pipe_class_compare);
%}
instruct cmpFastUnlock(flagsReg crx, iRegPdst oop, iRegPdst box, iRegPdst tmp1, iRegPdst tmp2, iRegPdst tmp3) %{
match(Set crx (FastUnlock oop box));
effect(TEMP tmp1, TEMP tmp2, TEMP tmp3);
predicate(!Compile::current()->use_rtm());
format %{ "FASTUNLOCK $oop, $box, $tmp1, $tmp2" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ compiler_fast_unlock_object($crx$$CondRegister, $oop$$Register, $box$$Register,
$tmp1$$Register, $tmp2$$Register, $tmp3$$Register,
UseBiasedLocking && !UseOptoBiasInlining,
false);
// If unlocking was successfull, crx should indicate 'EQ'.
// The compiler generates a branch to the runtime call to
// _complete_monitor_unlocking_Java for the case where crx is 'NE'.
%}
ins_pipe(pipe_class_compare);
%}
instruct cmpFastUnlock_tm(flagsReg crx, iRegPdst oop, iRegPdst box, iRegPdst tmp1, iRegPdst tmp2, iRegPdst tmp3) %{
match(Set crx (FastUnlock oop box));
effect(TEMP tmp1, TEMP tmp2, TEMP tmp3);
predicate(Compile::current()->use_rtm());
format %{ "FASTUNLOCK $oop, $box, $tmp1, $tmp2 (TM)" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ compiler_fast_unlock_object($crx$$CondRegister, $oop$$Register, $box$$Register,
$tmp1$$Register, $tmp2$$Register, $tmp3$$Register,
/*Biased Locking*/ false, /*TM*/ true);
// If unlocking was successfull, crx should indicate 'EQ'.
// The compiler generates a branch to the runtime call to
// _complete_monitor_unlocking_Java for the case where crx is 'NE'.
%}
ins_pipe(pipe_class_compare);
%}
// Align address.
instruct align_addr(iRegPdst dst, iRegPsrc src, immLnegpow2 mask) %{
match(Set dst (CastX2P (AndL (CastP2X src) mask)));
format %{ "ANDDI $dst, $src, $mask \t// next aligned address" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldicr);
__ clrrdi($dst$$Register, $src$$Register, log2_long((jlong)-$mask$$constant));
%}
ins_pipe(pipe_class_default);
%}
// Array size computation.
instruct array_size(iRegLdst dst, iRegPsrc end, iRegPsrc start) %{
match(Set dst (SubL (CastP2X end) (CastP2X start)));
format %{ "SUB $dst, $end, $start \t// array size in bytes" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_subf);
__ subf($dst$$Register, $start$$Register, $end$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Clear-array with dynamic array-size.
instruct inlineCallClearArray(rarg1RegL cnt, rarg2RegP base, Universe dummy, regCTR ctr) %{
match(Set dummy (ClearArray cnt base));
effect(USE_KILL cnt, USE_KILL base, KILL ctr);
ins_cost(MEMORY_REF_COST);
ins_alignment(8); // 'compute_padding()' gets called, up to this number-1 nops will get inserted.
format %{ "ClearArray $cnt, $base" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ clear_memory_doubleword($base$$Register, $cnt$$Register); // kills cnt, base, R0
%}
ins_pipe(pipe_class_default);
%}
// String_IndexOf for needle of length 1.
//
// Match needle into immediate operands: no loadConP node needed. Saves one
// register and two instructions over string_indexOf_imm1Node.
//
// Assumes register result differs from all input registers.
//
// Preserves registers haystack, haycnt
// Kills registers tmp1, tmp2
// Defines registers result
//
// Use dst register classes if register gets killed, as it is the case for tmp registers!
//
// Unfortunately this does not match too often. In many situations the AddP is used
// by several nodes, even several StrIndexOf nodes, breaking the match tree.
instruct string_indexOf_imm1_char(iRegIdst result, iRegPsrc haystack, iRegIsrc haycnt,
immP needleImm, immL offsetImm, immI_1 needlecntImm,
iRegIdst tmp1, iRegIdst tmp2,
flagsRegCR0 cr0, flagsRegCR1 cr1, regCTR ctr) %{
predicate(SpecialStringIndexOf && !CompactStrings); // type check implicit by parameter type, See Matcher::match_rule_supported
match(Set result (StrIndexOf (Binary haystack haycnt) (Binary (AddP needleImm offsetImm) needlecntImm)));
effect(TEMP_DEF result, TEMP tmp1, TEMP tmp2, KILL cr0, KILL cr1, KILL ctr);
ins_cost(150);
format %{ "String IndexOf CSCL1 $haystack[0..$haycnt], $needleImm+$offsetImm[0..$needlecntImm]"
"-> $result \t// KILL $haycnt, $tmp1, $tmp2, $cr0, $cr1" %}
ins_alignment(8); // 'compute_padding()' gets called, up to this number-1 nops will get inserted
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
immPOper *needleOper = (immPOper *)$needleImm;
const TypeOopPtr *t = needleOper->type()->isa_oopptr();
ciTypeArray* needle_values = t->const_oop()->as_type_array(); // Pointer to live char *
jchar chr;
if (java_lang_String::has_coder_field()) {
// New compact strings byte array strings
#ifdef VM_LITTLE_ENDIAN
chr = (((jchar)needle_values->element_value(1).as_byte()) << 8) |
(jchar)needle_values->element_value(0).as_byte();
#else
chr = (((jchar)needle_values->element_value(0).as_byte()) << 8) |
(jchar)needle_values->element_value(1).as_byte();
#endif
} else {
// Old char array strings
chr = needle_values->char_at(0);
}
__ string_indexof_1($result$$Register,
$haystack$$Register, $haycnt$$Register,
R0, chr,
$tmp1$$Register, $tmp2$$Register);
%}
ins_pipe(pipe_class_compare);
%}
// String_IndexOf for needle of length 1.
//
// Special case requires less registers and emits less instructions.
//
// Assumes register result differs from all input registers.
//
// Preserves registers haystack, haycnt
// Kills registers tmp1, tmp2, needle
// Defines registers result
//
// Use dst register classes if register gets killed, as it is the case for tmp registers!
instruct string_indexOf_imm1(iRegIdst result, iRegPsrc haystack, iRegIsrc haycnt,
rscratch2RegP needle, immI_1 needlecntImm,
iRegIdst tmp1, iRegIdst tmp2,
flagsRegCR0 cr0, flagsRegCR1 cr1, regCTR ctr) %{
match(Set result (StrIndexOf (Binary haystack haycnt) (Binary needle needlecntImm)));
effect(USE_KILL needle, /* TDEF needle, */ TEMP_DEF result,
TEMP tmp1, TEMP tmp2, KILL cr0, KILL cr1, KILL ctr);
// Required for EA: check if it is still a type_array.
predicate(SpecialStringIndexOf && !CompactStrings &&
n->in(3)->in(1)->bottom_type()->is_aryptr()->const_oop() &&
n->in(3)->in(1)->bottom_type()->is_aryptr()->const_oop()->is_type_array());
ins_cost(180);
ins_alignment(8); // 'compute_padding()' gets called, up to this number-1 nops will get inserted.
format %{ "String IndexOf SCL1 $haystack[0..$haycnt], $needle[0..$needlecntImm]"
" -> $result \t// KILL $haycnt, $needle, $tmp1, $tmp2, $cr0, $cr1" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
Node *ndl = in(operand_index($needle)); // The node that defines needle.
ciTypeArray* needle_values = ndl->bottom_type()->is_aryptr()->const_oop()->as_type_array();
guarantee(needle_values, "sanity");
jchar chr;
if (java_lang_String::has_coder_field()) {
// New compact strings byte array strings
#ifdef VM_LITTLE_ENDIAN
chr = (((jchar)needle_values->element_value(1).as_byte()) << 8) |
(jchar)needle_values->element_value(0).as_byte();
#else
chr = (((jchar)needle_values->element_value(0).as_byte()) << 8) |
(jchar)needle_values->element_value(1).as_byte();
#endif
} else {
// Old char array strings
chr = needle_values->char_at(0);
}
__ string_indexof_1($result$$Register,
$haystack$$Register, $haycnt$$Register,
R0, chr,
$tmp1$$Register, $tmp2$$Register);
%}
ins_pipe(pipe_class_compare);
%}
// String_IndexOfChar
//
// Assumes register result differs from all input registers.
//
// Preserves registers haystack, haycnt
// Kills registers tmp1, tmp2
// Defines registers result
//
// Use dst register classes if register gets killed, as it is the case for tmp registers!
instruct string_indexOfChar(iRegIdst result, iRegPsrc haystack, iRegIsrc haycnt,
iRegIsrc ch, iRegIdst tmp1, iRegIdst tmp2,
flagsRegCR0 cr0, flagsRegCR1 cr1, regCTR ctr) %{
match(Set result (StrIndexOfChar (Binary haystack haycnt) ch));
effect(TEMP_DEF result, TEMP tmp1, TEMP tmp2, KILL cr0, KILL cr1, KILL ctr);
predicate(SpecialStringIndexOf && !CompactStrings);
ins_cost(180);
ins_alignment(8); // 'compute_padding()' gets called, up to this number-1 nops will get inserted.
format %{ "String IndexOfChar $haystack[0..$haycnt], $ch"
" -> $result \t// KILL $haycnt, $tmp1, $tmp2, $cr0, $cr1" %}
ins_encode %{
__ string_indexof_1($result$$Register,
$haystack$$Register, $haycnt$$Register,
$ch$$Register, 0 /* this is not used if the character is already in a register */,
$tmp1$$Register, $tmp2$$Register);
%}
ins_pipe(pipe_class_compare);
%}
// String_IndexOf.
//
// Length of needle as immediate. This saves instruction loading constant needle
// length.
// @@@ TODO Specify rules for length < 8 or so, and roll out comparison of needle
// completely or do it in vector instruction. This should save registers for
// needlecnt and needle.
//
// Assumes register result differs from all input registers.
// Overwrites haycnt, needlecnt.
// Use dst register classes if register gets killed, as it is the case for tmp registers!
instruct string_indexOf_imm(iRegIdst result, iRegPsrc haystack, rscratch1RegI haycnt,
iRegPsrc needle, uimmI15 needlecntImm,
iRegIdst tmp1, iRegIdst tmp2, iRegIdst tmp3, iRegIdst tmp4, iRegIdst tmp5,
flagsRegCR0 cr0, flagsRegCR1 cr1, flagsRegCR6 cr6, regCTR ctr) %{
match(Set result (StrIndexOf (Binary haystack haycnt) (Binary needle needlecntImm)));
effect(USE_KILL haycnt, /* better: TDEF haycnt, */ TEMP_DEF result,
TEMP tmp1, TEMP tmp2, TEMP tmp3, TEMP tmp4, TEMP tmp5, KILL cr0, KILL cr1, KILL cr6, KILL ctr);
// Required for EA: check if it is still a type_array.
predicate(SpecialStringIndexOf && !CompactStrings && n->in(3)->in(1)->bottom_type()->is_aryptr()->const_oop() &&
n->in(3)->in(1)->bottom_type()->is_aryptr()->const_oop()->is_type_array());
ins_cost(250);
ins_alignment(8); // 'compute_padding()' gets called, up to this number-1 nops will get inserted.
format %{ "String IndexOf SCL $haystack[0..$haycnt], $needle[0..$needlecntImm]"
" -> $result \t// KILL $haycnt, $tmp1, $tmp2, $tmp3, $tmp4, $tmp5, $cr0, $cr1" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
Node *ndl = in(operand_index($needle)); // The node that defines needle.
ciTypeArray* needle_values = ndl->bottom_type()->is_aryptr()->const_oop()->as_type_array();
__ string_indexof($result$$Register,
$haystack$$Register, $haycnt$$Register,
$needle$$Register, needle_values, $tmp5$$Register, $needlecntImm$$constant,
$tmp1$$Register, $tmp2$$Register, $tmp3$$Register, $tmp4$$Register);
%}
ins_pipe(pipe_class_compare);
%}
// StrIndexOf node.
//
// Assumes register result differs from all input registers.
// Overwrites haycnt, needlecnt.
// Use dst register classes if register gets killed, as it is the case for tmp registers!
instruct string_indexOf(iRegIdst result, iRegPsrc haystack, rscratch1RegI haycnt, iRegPsrc needle, rscratch2RegI needlecnt,
iRegLdst tmp1, iRegLdst tmp2, iRegLdst tmp3, iRegLdst tmp4,
flagsRegCR0 cr0, flagsRegCR1 cr1, flagsRegCR6 cr6, regCTR ctr) %{
match(Set result (StrIndexOf (Binary haystack haycnt) (Binary needle needlecnt)));
effect(USE_KILL haycnt, USE_KILL needlecnt, /*better: TDEF haycnt, TDEF needlecnt,*/
TEMP_DEF result,
TEMP tmp1, TEMP tmp2, TEMP tmp3, TEMP tmp4, KILL cr0, KILL cr1, KILL cr6, KILL ctr);
predicate(SpecialStringIndexOf && !CompactStrings); // See Matcher::match_rule_supported.
ins_cost(300);
ins_alignment(8); // 'compute_padding()' gets called, up to this number-1 nops will get inserted.
format %{ "String IndexOf $haystack[0..$haycnt], $needle[0..$needlecnt]"
" -> $result \t// KILL $haycnt, $needlecnt, $tmp1, $tmp2, $tmp3, $tmp4, $cr0, $cr1" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ string_indexof($result$$Register,
$haystack$$Register, $haycnt$$Register,
$needle$$Register, NULL, $needlecnt$$Register, 0, // needlecnt not constant.
$tmp1$$Register, $tmp2$$Register, $tmp3$$Register, $tmp4$$Register);
%}
ins_pipe(pipe_class_compare);
%}
// String equals with immediate.
instruct string_equals_imm(iRegPsrc str1, iRegPsrc str2, uimmI15 cntImm, iRegIdst result,
iRegPdst tmp1, iRegPdst tmp2,
flagsRegCR0 cr0, flagsRegCR6 cr6, regCTR ctr) %{
match(Set result (StrEquals (Binary str1 str2) cntImm));
effect(TEMP_DEF result, TEMP tmp1, TEMP tmp2,
KILL cr0, KILL cr6, KILL ctr);
predicate(SpecialStringEquals && !CompactStrings); // See Matcher::match_rule_supported.
ins_cost(250);
ins_alignment(8); // 'compute_padding()' gets called, up to this number-1 nops will get inserted.
format %{ "String Equals SCL [0..$cntImm]($str1),[0..$cntImm]($str2)"
" -> $result \t// KILL $cr0, $cr6, $ctr, TEMP $result, $tmp1, $tmp2" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ char_arrays_equalsImm($str1$$Register, $str2$$Register, $cntImm$$constant,
$result$$Register, $tmp1$$Register, $tmp2$$Register);
%}
ins_pipe(pipe_class_compare);
%}
// String equals.
// Use dst register classes if register gets killed, as it is the case for TEMP operands!
instruct string_equals(iRegPsrc str1, iRegPsrc str2, iRegIsrc cnt, iRegIdst result,
iRegPdst tmp1, iRegPdst tmp2, iRegPdst tmp3, iRegPdst tmp4, iRegPdst tmp5,
flagsRegCR0 cr0, flagsRegCR1 cr1, flagsRegCR6 cr6, regCTR ctr) %{
match(Set result (StrEquals (Binary str1 str2) cnt));
effect(TEMP_DEF result, TEMP tmp1, TEMP tmp2, TEMP tmp3, TEMP tmp4, TEMP tmp5,
KILL cr0, KILL cr1, KILL cr6, KILL ctr);
predicate(SpecialStringEquals && !CompactStrings); // See Matcher::match_rule_supported.
ins_cost(300);
ins_alignment(8); // 'compute_padding()' gets called, up to this number-1 nops will get inserted.
format %{ "String Equals [0..$cnt]($str1),[0..$cnt]($str2) -> $result"
" \t// KILL $cr0, $cr1, $cr6, $ctr, TEMP $result, $tmp1, $tmp2, $tmp3, $tmp4, $tmp5" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ char_arrays_equals($str1$$Register, $str2$$Register, $cnt$$Register, $result$$Register,
$tmp1$$Register, $tmp2$$Register, $tmp3$$Register, $tmp4$$Register, $tmp5$$Register);
%}
ins_pipe(pipe_class_compare);
%}
// String compare.
// Char[] pointers are passed in.
// Use dst register classes if register gets killed, as it is the case for TEMP operands!
instruct string_compare(rarg1RegP str1, rarg2RegP str2, rarg3RegI cnt1, rarg4RegI cnt2, iRegIdst result,
iRegPdst tmp, flagsRegCR0 cr0, regCTR ctr) %{
predicate(!CompactStrings);
match(Set result (StrComp (Binary str1 cnt1) (Binary str2 cnt2)));
effect(USE_KILL cnt1, USE_KILL cnt2, USE_KILL str1, USE_KILL str2, TEMP_DEF result, TEMP tmp, KILL cr0, KILL ctr);
ins_cost(300);
ins_alignment(8); // 'compute_padding()' gets called, up to this number-1 nops will get inserted.
format %{ "String Compare $str1[0..$cnt1], $str2[0..$cnt2] -> $result"
" \t// TEMP $tmp, $result KILLs $str1, $cnt1, $str2, $cnt2, $cr0, $ctr" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ string_compare($str1$$Register, $str2$$Register, $cnt1$$Register, $cnt2$$Register,
$result$$Register, $tmp$$Register);
%}
ins_pipe(pipe_class_compare);
%}
//---------- Min/Max Instructions ---------------------------------------------
instruct minI_reg_reg_Ex(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
match(Set dst (MinI src1 src2));
ins_cost(DEFAULT_COST*6);
expand %{
iRegLdst src1s;
iRegLdst src2s;
iRegLdst diff;
iRegLdst sm;
iRegLdst doz; // difference or zero
convI2L_reg(src1s, src1); // Ensure proper sign extension.
convI2L_reg(src2s, src2); // Ensure proper sign extension.
subL_reg_reg(diff, src2s, src1s);
// Need to consider >=33 bit result, therefore we need signmaskL.
signmask64L_regL(sm, diff);
andL_reg_reg(doz, diff, sm); // <=0
addI_regL_regL(dst, doz, src1s);
%}
%}
instruct maxI_reg_reg_Ex(iRegIdst dst, iRegIsrc src1, iRegIsrc src2) %{
match(Set dst (MaxI src1 src2));
ins_cost(DEFAULT_COST*6);
expand %{
iRegLdst src1s;
iRegLdst src2s;
iRegLdst diff;
iRegLdst sm;
iRegLdst doz; // difference or zero
convI2L_reg(src1s, src1); // Ensure proper sign extension.
convI2L_reg(src2s, src2); // Ensure proper sign extension.
subL_reg_reg(diff, src2s, src1s);
// Need to consider >=33 bit result, therefore we need signmaskL.
signmask64L_regL(sm, diff);
andcL_reg_reg(doz, diff, sm); // >=0
addI_regL_regL(dst, doz, src1s);
%}
%}
//---------- Population Count Instructions ------------------------------------
// Popcnt for Power7.
instruct popCountI(iRegIdst dst, iRegIsrc src) %{
match(Set dst (PopCountI src));
predicate(UsePopCountInstruction && VM_Version::has_popcntw());
ins_cost(DEFAULT_COST);
format %{ "POPCNTW $dst, $src" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_popcntb);
__ popcntw($dst$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
// Popcnt for Power7.
instruct popCountL(iRegIdst dst, iRegLsrc src) %{
predicate(UsePopCountInstruction && VM_Version::has_popcntw());
match(Set dst (PopCountL src));
ins_cost(DEFAULT_COST);
format %{ "POPCNTD $dst, $src" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_popcntb);
__ popcntd($dst$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct countLeadingZerosI(iRegIdst dst, iRegIsrc src) %{
match(Set dst (CountLeadingZerosI src));
predicate(UseCountLeadingZerosInstructionsPPC64); // See Matcher::match_rule_supported.
ins_cost(DEFAULT_COST);
format %{ "CNTLZW $dst, $src" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_cntlzw);
__ cntlzw($dst$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct countLeadingZerosL(iRegIdst dst, iRegLsrc src) %{
match(Set dst (CountLeadingZerosL src));
predicate(UseCountLeadingZerosInstructionsPPC64); // See Matcher::match_rule_supported.
ins_cost(DEFAULT_COST);
format %{ "CNTLZD $dst, $src" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_cntlzd);
__ cntlzd($dst$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct countLeadingZerosP(iRegIdst dst, iRegPsrc src) %{
// no match-rule, false predicate
effect(DEF dst, USE src);
predicate(false);
format %{ "CNTLZD $dst, $src" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_cntlzd);
__ cntlzd($dst$$Register, $src$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct countTrailingZerosI_Ex(iRegIdst dst, iRegIsrc src) %{
match(Set dst (CountTrailingZerosI src));
predicate(UseCountLeadingZerosInstructionsPPC64);
ins_cost(DEFAULT_COST);
expand %{
immI16 imm1 %{ (int)-1 %}
immI16 imm2 %{ (int)32 %}
immI_minus1 m1 %{ -1 %}
iRegIdst tmpI1;
iRegIdst tmpI2;
iRegIdst tmpI3;
addI_reg_imm16(tmpI1, src, imm1);
andcI_reg_reg(tmpI2, src, m1, tmpI1);
countLeadingZerosI(tmpI3, tmpI2);
subI_imm16_reg(dst, imm2, tmpI3);
%}
%}
instruct countTrailingZerosL_Ex(iRegIdst dst, iRegLsrc src) %{
match(Set dst (CountTrailingZerosL src));
predicate(UseCountLeadingZerosInstructionsPPC64);
ins_cost(DEFAULT_COST);
expand %{
immL16 imm1 %{ (long)-1 %}
immI16 imm2 %{ (int)64 %}
iRegLdst tmpL1;
iRegLdst tmpL2;
iRegIdst tmpL3;
addL_reg_imm16(tmpL1, src, imm1);
andcL_reg_reg(tmpL2, tmpL1, src);
countLeadingZerosL(tmpL3, tmpL2);
subI_imm16_reg(dst, imm2, tmpL3);
%}
%}
// Expand nodes for byte_reverse_int.
instruct insrwi_a(iRegIdst dst, iRegIsrc src, immI16 pos, immI16 shift) %{
effect(DEF dst, USE src, USE pos, USE shift);
predicate(false);
format %{ "INSRWI $dst, $src, $pos, $shift" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rlwimi);
__ insrwi($dst$$Register, $src$$Register, $shift$$constant, $pos$$constant);
%}
ins_pipe(pipe_class_default);
%}
// As insrwi_a, but with USE_DEF.
instruct insrwi(iRegIdst dst, iRegIsrc src, immI16 pos, immI16 shift) %{
effect(USE_DEF dst, USE src, USE pos, USE shift);
predicate(false);
format %{ "INSRWI $dst, $src, $pos, $shift" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rlwimi);
__ insrwi($dst$$Register, $src$$Register, $shift$$constant, $pos$$constant);
%}
ins_pipe(pipe_class_default);
%}
// Just slightly faster than java implementation.
instruct bytes_reverse_int_Ex(iRegIdst dst, iRegIsrc src) %{
match(Set dst (ReverseBytesI src));
predicate(UseCountLeadingZerosInstructionsPPC64);
ins_cost(DEFAULT_COST);
expand %{
immI16 imm24 %{ (int) 24 %}
immI16 imm16 %{ (int) 16 %}
immI16 imm8 %{ (int) 8 %}
immI16 imm4 %{ (int) 4 %}
immI16 imm0 %{ (int) 0 %}
iRegLdst tmpI1;
iRegLdst tmpI2;
iRegLdst tmpI3;
urShiftI_reg_imm(tmpI1, src, imm24);
insrwi_a(dst, tmpI1, imm24, imm8);
urShiftI_reg_imm(tmpI2, src, imm16);
insrwi(dst, tmpI2, imm8, imm16);
urShiftI_reg_imm(tmpI3, src, imm8);
insrwi(dst, tmpI3, imm8, imm8);
insrwi(dst, src, imm0, imm8);
%}
%}
//---------- Replicate Vector Instructions ------------------------------------
// Insrdi does replicate if src == dst.
instruct repl32(iRegLdst dst) %{
predicate(false);
effect(USE_DEF dst);
format %{ "INSRDI $dst, #0, $dst, #32 \t// replicate" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldimi);
__ insrdi($dst$$Register, $dst$$Register, 32, 0);
%}
ins_pipe(pipe_class_default);
%}
// Insrdi does replicate if src == dst.
instruct repl48(iRegLdst dst) %{
predicate(false);
effect(USE_DEF dst);
format %{ "INSRDI $dst, #0, $dst, #48 \t// replicate" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldimi);
__ insrdi($dst$$Register, $dst$$Register, 48, 0);
%}
ins_pipe(pipe_class_default);
%}
// Insrdi does replicate if src == dst.
instruct repl56(iRegLdst dst) %{
predicate(false);
effect(USE_DEF dst);
format %{ "INSRDI $dst, #0, $dst, #56 \t// replicate" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_rldimi);
__ insrdi($dst$$Register, $dst$$Register, 56, 0);
%}
ins_pipe(pipe_class_default);
%}
instruct repl8B_reg_Ex(iRegLdst dst, iRegIsrc src) %{
match(Set dst (ReplicateB src));
predicate(n->as_Vector()->length() == 8);
expand %{
moveReg(dst, src);
repl56(dst);
repl48(dst);
repl32(dst);
%}
%}
instruct repl8B_immI0(iRegLdst dst, immI_0 zero) %{
match(Set dst (ReplicateB zero));
predicate(n->as_Vector()->length() == 8);
format %{ "LI $dst, #0 \t// replicate8B" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addi);
__ li($dst$$Register, (int)((short)($zero$$constant & 0xFFFF)));
%}
ins_pipe(pipe_class_default);
%}
instruct repl8B_immIminus1(iRegLdst dst, immI_minus1 src) %{
match(Set dst (ReplicateB src));
predicate(n->as_Vector()->length() == 8);
format %{ "LI $dst, #-1 \t// replicate8B" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addi);
__ li($dst$$Register, (int)((short)($src$$constant & 0xFFFF)));
%}
ins_pipe(pipe_class_default);
%}
instruct repl4S_reg_Ex(iRegLdst dst, iRegIsrc src) %{
match(Set dst (ReplicateS src));
predicate(n->as_Vector()->length() == 4);
expand %{
moveReg(dst, src);
repl48(dst);
repl32(dst);
%}
%}
instruct repl4S_immI0(iRegLdst dst, immI_0 zero) %{
match(Set dst (ReplicateS zero));
predicate(n->as_Vector()->length() == 4);
format %{ "LI $dst, #0 \t// replicate4C" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addi);
__ li($dst$$Register, (int)((short)($zero$$constant & 0xFFFF)));
%}
ins_pipe(pipe_class_default);
%}
instruct repl4S_immIminus1(iRegLdst dst, immI_minus1 src) %{
match(Set dst (ReplicateS src));
predicate(n->as_Vector()->length() == 4);
format %{ "LI $dst, -1 \t// replicate4C" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addi);
__ li($dst$$Register, (int)((short)($src$$constant & 0xFFFF)));
%}
ins_pipe(pipe_class_default);
%}
instruct repl2I_reg_Ex(iRegLdst dst, iRegIsrc src) %{
match(Set dst (ReplicateI src));
predicate(n->as_Vector()->length() == 2);
ins_cost(2 * DEFAULT_COST);
expand %{
moveReg(dst, src);
repl32(dst);
%}
%}
instruct repl2I_immI0(iRegLdst dst, immI_0 zero) %{
match(Set dst (ReplicateI zero));
predicate(n->as_Vector()->length() == 2);
format %{ "LI $dst, #0 \t// replicate4C" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addi);
__ li($dst$$Register, (int)((short)($zero$$constant & 0xFFFF)));
%}
ins_pipe(pipe_class_default);
%}
instruct repl2I_immIminus1(iRegLdst dst, immI_minus1 src) %{
match(Set dst (ReplicateI src));
predicate(n->as_Vector()->length() == 2);
format %{ "LI $dst, -1 \t// replicate4C" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addi);
__ li($dst$$Register, (int)((short)($src$$constant & 0xFFFF)));
%}
ins_pipe(pipe_class_default);
%}
// Move float to int register via stack, replicate.
instruct repl2F_reg_Ex(iRegLdst dst, regF src) %{
match(Set dst (ReplicateF src));
predicate(n->as_Vector()->length() == 2);
ins_cost(2 * MEMORY_REF_COST + DEFAULT_COST);
expand %{
stackSlotL tmpS;
iRegIdst tmpI;
moveF2I_reg_stack(tmpS, src); // Move float to stack.
moveF2I_stack_reg(tmpI, tmpS); // Move stack to int reg.
moveReg(dst, tmpI); // Move int to long reg.
repl32(dst); // Replicate bitpattern.
%}
%}
// Replicate scalar constant to packed float values in Double register
instruct repl2F_immF_Ex(iRegLdst dst, immF src) %{
match(Set dst (ReplicateF src));
predicate(n->as_Vector()->length() == 2);
ins_cost(5 * DEFAULT_COST);
format %{ "LD $dst, offset, $constanttablebase\t// load replicated float $src $src from table, postalloc expanded" %}
postalloc_expand( postalloc_expand_load_replF_constant(dst, src, constanttablebase) );
%}
// Replicate scalar zero constant to packed float values in Double register
instruct repl2F_immF0(iRegLdst dst, immF_0 zero) %{
match(Set dst (ReplicateF zero));
predicate(n->as_Vector()->length() == 2);
format %{ "LI $dst, #0 \t// replicate2F" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_addi);
__ li($dst$$Register, 0x0);
%}
ins_pipe(pipe_class_default);
%}
//----------Overflow Math Instructions-----------------------------------------
// Note that we have to make sure that XER.SO is reset before using overflow instructions.
// Simple Overflow operations can be matched by very few instructions (e.g. addExact: xor, and_, bc).
// Seems like only Long intrinsincs have an advantage. (The only expensive one is OverflowMulL.)
instruct overflowAddL_reg_reg(flagsRegCR0 cr0, iRegLsrc op1, iRegLsrc op2) %{
match(Set cr0 (OverflowAddL op1 op2));
format %{ "add_ $op1, $op2\t# overflow check long" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ li(R0, 0);
__ mtxer(R0); // clear XER.SO
__ addo_(R0, $op1$$Register, $op2$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct overflowSubL_reg_reg(flagsRegCR0 cr0, iRegLsrc op1, iRegLsrc op2) %{
match(Set cr0 (OverflowSubL op1 op2));
format %{ "subfo_ R0, $op2, $op1\t# overflow check long" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ li(R0, 0);
__ mtxer(R0); // clear XER.SO
__ subfo_(R0, $op2$$Register, $op1$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct overflowNegL_reg(flagsRegCR0 cr0, immL_0 zero, iRegLsrc op2) %{
match(Set cr0 (OverflowSubL zero op2));
format %{ "nego_ R0, $op2\t# overflow check long" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ li(R0, 0);
__ mtxer(R0); // clear XER.SO
__ nego_(R0, $op2$$Register);
%}
ins_pipe(pipe_class_default);
%}
instruct overflowMulL_reg_reg(flagsRegCR0 cr0, iRegLsrc op1, iRegLsrc op2) %{
match(Set cr0 (OverflowMulL op1 op2));
format %{ "mulldo_ R0, $op1, $op2\t# overflow check long" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ li(R0, 0);
__ mtxer(R0); // clear XER.SO
__ mulldo_(R0, $op1$$Register, $op2$$Register);
%}
ins_pipe(pipe_class_default);
%}
// ============================================================================
// Safepoint Instruction
instruct safePoint_poll(iRegPdst poll) %{
match(SafePoint poll);
predicate(LoadPollAddressFromThread);
// It caused problems to add the effect that r0 is killed, but this
// effect no longer needs to be mentioned, since r0 is not contained
// in a reg_class.
format %{ "LD R0, #0, $poll \t// Safepoint poll for GC" %}
size(4);
ins_encode( enc_poll(0x0, poll) );
ins_pipe(pipe_class_default);
%}
// Safepoint without per-thread support. Load address of page to poll
// as constant.
// Rscratch2RegP is R12.
// LoadConPollAddr node is added in pd_post_matching_hook(). It must be
// a seperate node so that the oop map is at the right location.
instruct safePoint_poll_conPollAddr(rscratch2RegP poll) %{
match(SafePoint poll);
predicate(!LoadPollAddressFromThread);
// It caused problems to add the effect that r0 is killed, but this
// effect no longer needs to be mentioned, since r0 is not contained
// in a reg_class.
format %{ "LD R0, #0, R12 \t// Safepoint poll for GC" %}
ins_encode( enc_poll(0x0, poll) );
ins_pipe(pipe_class_default);
%}
// ============================================================================
// Call Instructions
// Call Java Static Instruction
// Schedulable version of call static node.
instruct CallStaticJavaDirect(method meth) %{
match(CallStaticJava);
effect(USE meth);
ins_cost(CALL_COST);
ins_num_consts(3 /* up to 3 patchable constants: inline cache, 2 call targets. */);
format %{ "CALL,static $meth \t// ==> " %}
size(4);
ins_encode( enc_java_static_call(meth) );
ins_pipe(pipe_class_call);
%}
// Call Java Dynamic Instruction
// Used by postalloc expand of CallDynamicJavaDirectSchedEx (actual call).
// Loading of IC was postalloc expanded. The nodes loading the IC are reachable
// via fields ins_field_load_ic_hi_node and ins_field_load_ic_node.
// The call destination must still be placed in the constant pool.
instruct CallDynamicJavaDirectSched(method meth) %{
match(CallDynamicJava); // To get all the data fields we need ...
effect(USE meth);
predicate(false); // ... but never match.
ins_field_load_ic_hi_node(loadConL_hiNode*);
ins_field_load_ic_node(loadConLNode*);
ins_num_consts(1 /* 1 patchable constant: call destination */);
format %{ "BL \t// dynamic $meth ==> " %}
size(4);
ins_encode( enc_java_dynamic_call_sched(meth) );
ins_pipe(pipe_class_call);
%}
// Schedulable (i.e. postalloc expanded) version of call dynamic java.
// We use postalloc expanded calls if we use inline caches
// and do not update method data.
//
// This instruction has two constants: inline cache (IC) and call destination.
// Loading the inline cache will be postalloc expanded, thus leaving a call with
// one constant.
instruct CallDynamicJavaDirectSched_Ex(method meth) %{
match(CallDynamicJava);
effect(USE meth);
predicate(UseInlineCaches);
ins_cost(CALL_COST);
ins_num_consts(2 /* 2 patchable constants: inline cache, call destination. */);
format %{ "CALL,dynamic $meth \t// postalloc expanded" %}
postalloc_expand( postalloc_expand_java_dynamic_call_sched(meth, constanttablebase) );
%}
// Compound version of call dynamic java
// We use postalloc expanded calls if we use inline caches
// and do not update method data.
instruct CallDynamicJavaDirect(method meth) %{
match(CallDynamicJava);
effect(USE meth);
predicate(!UseInlineCaches);
ins_cost(CALL_COST);
// Enc_java_to_runtime_call needs up to 4 constants (method data oop).
ins_num_consts(4);
format %{ "CALL,dynamic $meth \t// ==> " %}
ins_encode( enc_java_dynamic_call(meth, constanttablebase) );
ins_pipe(pipe_class_call);
%}
// Call Runtime Instruction
instruct CallRuntimeDirect(method meth) %{
match(CallRuntime);
effect(USE meth);
ins_cost(CALL_COST);
// Enc_java_to_runtime_call needs up to 3 constants: call target,
// env for callee, C-toc.
ins_num_consts(3);
format %{ "CALL,runtime" %}
ins_encode( enc_java_to_runtime_call(meth) );
ins_pipe(pipe_class_call);
%}
// Call Leaf
// Used by postalloc expand of CallLeafDirect_Ex (mtctr).
instruct CallLeafDirect_mtctr(iRegLdst dst, iRegLsrc src) %{
effect(DEF dst, USE src);
ins_num_consts(1);
format %{ "MTCTR $src" %}
size(4);
ins_encode( enc_leaf_call_mtctr(src) );
ins_pipe(pipe_class_default);
%}
// Used by postalloc expand of CallLeafDirect_Ex (actual call).
instruct CallLeafDirect(method meth) %{
match(CallLeaf); // To get the data all the data fields we need ...
effect(USE meth);
predicate(false); // but never match.
format %{ "BCTRL \t// leaf call $meth ==> " %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_bctrl);
__ bctrl();
%}
ins_pipe(pipe_class_call);
%}
// postalloc expand of CallLeafDirect.
// Load adress to call from TOC, then bl to it.
instruct CallLeafDirect_Ex(method meth) %{
match(CallLeaf);
effect(USE meth);
ins_cost(CALL_COST);
// Postalloc_expand_java_to_runtime_call needs up to 3 constants: call target,
// env for callee, C-toc.
ins_num_consts(3);
format %{ "CALL,runtime leaf $meth \t// postalloc expanded" %}
postalloc_expand( postalloc_expand_java_to_runtime_call(meth, constanttablebase) );
%}
// Call runtime without safepoint - same as CallLeaf.
// postalloc expand of CallLeafNoFPDirect.
// Load adress to call from TOC, then bl to it.
instruct CallLeafNoFPDirect_Ex(method meth) %{
match(CallLeafNoFP);
effect(USE meth);
ins_cost(CALL_COST);
// Enc_java_to_runtime_call needs up to 3 constants: call target,
// env for callee, C-toc.
ins_num_consts(3);
format %{ "CALL,runtime leaf nofp $meth \t// postalloc expanded" %}
postalloc_expand( postalloc_expand_java_to_runtime_call(meth, constanttablebase) );
%}
// Tail Call; Jump from runtime stub to Java code.
// Also known as an 'interprocedural jump'.
// Target of jump will eventually return to caller.
// TailJump below removes the return address.
instruct TailCalljmpInd(iRegPdstNoScratch jump_target, inline_cache_regP method_oop) %{
match(TailCall jump_target method_oop);
ins_cost(CALL_COST);
format %{ "MTCTR $jump_target \t// $method_oop holds method oop\n\t"
"BCTR \t// tail call" %}
size(8);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ mtctr($jump_target$$Register);
__ bctr();
%}
ins_pipe(pipe_class_call);
%}
// Return Instruction
instruct Ret() %{
match(Return);
format %{ "BLR \t// branch to link register" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_blr);
// LR is restored in MachEpilogNode. Just do the RET here.
__ blr();
%}
ins_pipe(pipe_class_default);
%}
// Tail Jump; remove the return address; jump to target.
// TailCall above leaves the return address around.
// TailJump is used in only one place, the rethrow_Java stub (fancy_jump=2).
// ex_oop (Exception Oop) is needed in %o0 at the jump. As there would be a
// "restore" before this instruction (in Epilogue), we need to materialize it
// in %i0.
instruct tailjmpInd(iRegPdstNoScratch jump_target, rarg1RegP ex_oop) %{
match(TailJump jump_target ex_oop);
ins_cost(CALL_COST);
format %{ "LD R4_ARG2 = LR\n\t"
"MTCTR $jump_target\n\t"
"BCTR \t// TailJump, exception oop: $ex_oop" %}
size(12);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
__ ld(R4_ARG2/* issuing pc */, _abi(lr), R1_SP);
__ mtctr($jump_target$$Register);
__ bctr();
%}
ins_pipe(pipe_class_call);
%}
// Create exception oop: created by stack-crawling runtime code.
// Created exception is now available to this handler, and is setup
// just prior to jumping to this handler. No code emitted.
instruct CreateException(rarg1RegP ex_oop) %{
match(Set ex_oop (CreateEx));
ins_cost(0);
format %{ " -- \t// exception oop; no code emitted" %}
size(0);
ins_encode( /*empty*/ );
ins_pipe(pipe_class_default);
%}
// Rethrow exception: The exception oop will come in the first
// argument position. Then JUMP (not call) to the rethrow stub code.
instruct RethrowException() %{
match(Rethrow);
ins_cost(CALL_COST);
format %{ "Jmp rethrow_stub" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_compound);
cbuf.set_insts_mark();
__ b64_patchable((address)OptoRuntime::rethrow_stub(), relocInfo::runtime_call_type);
%}
ins_pipe(pipe_class_call);
%}
// Die now.
instruct ShouldNotReachHere() %{
match(Halt);
ins_cost(CALL_COST);
format %{ "ShouldNotReachHere" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_tdi);
__ trap_should_not_reach_here();
%}
ins_pipe(pipe_class_default);
%}
// This name is KNOWN by the ADLC and cannot be changed. The ADLC
// forces a 'TypeRawPtr::BOTTOM' output type for this guy.
// Get a DEF on threadRegP, no costs, no encoding, use
// 'ins_should_rematerialize(true)' to avoid spilling.
instruct tlsLoadP(threadRegP dst) %{
match(Set dst (ThreadLocal));
ins_cost(0);
ins_should_rematerialize(true);
format %{ " -- \t// $dst=Thread::current(), empty" %}
size(0);
ins_encode( /*empty*/ );
ins_pipe(pipe_class_empty);
%}
//---Some PPC specific nodes---------------------------------------------------
// Stop a group.
instruct endGroup() %{
ins_cost(0);
ins_is_nop(true);
format %{ "End Bundle (ori r1, r1, 0)" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_endgroup);
__ endgroup();
%}
ins_pipe(pipe_class_default);
%}
// Nop instructions
instruct fxNop() %{
ins_cost(0);
ins_is_nop(true);
format %{ "fxNop" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fmr);
__ nop();
%}
ins_pipe(pipe_class_default);
%}
instruct fpNop0() %{
ins_cost(0);
ins_is_nop(true);
format %{ "fpNop0" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fmr);
__ fpnop0();
%}
ins_pipe(pipe_class_default);
%}
instruct fpNop1() %{
ins_cost(0);
ins_is_nop(true);
format %{ "fpNop1" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_fmr);
__ fpnop1();
%}
ins_pipe(pipe_class_default);
%}
instruct brNop0() %{
ins_cost(0);
size(4);
format %{ "brNop0" %}
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_mcrf);
__ brnop0();
%}
ins_is_nop(true);
ins_pipe(pipe_class_default);
%}
instruct brNop1() %{
ins_cost(0);
ins_is_nop(true);
format %{ "brNop1" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_mcrf);
__ brnop1();
%}
ins_pipe(pipe_class_default);
%}
instruct brNop2() %{
ins_cost(0);
ins_is_nop(true);
format %{ "brNop2" %}
size(4);
ins_encode %{
// TODO: PPC port $archOpcode(ppc64Opcode_mcrf);
__ brnop2();
%}
ins_pipe(pipe_class_default);
%}
//----------PEEPHOLE RULES-----------------------------------------------------
// These must follow all instruction definitions as they use the names
// defined in the instructions definitions.
//
// peepmatch ( root_instr_name [preceeding_instruction]* );
//
// peepconstraint %{
// (instruction_number.operand_name relational_op instruction_number.operand_name
// [, ...] );
// // instruction numbers are zero-based using left to right order in peepmatch
//
// peepreplace ( instr_name ( [instruction_number.operand_name]* ) );
// // provide an instruction_number.operand_name for each operand that appears
// // in the replacement instruction's match rule
//
// ---------VM FLAGS---------------------------------------------------------
//
// All peephole optimizations can be turned off using -XX:-OptoPeephole
//
// Each peephole rule is given an identifying number starting with zero and
// increasing by one in the order seen by the parser. An individual peephole
// can be enabled, and all others disabled, by using -XX:OptoPeepholeAt=#
// on the command-line.
//
// ---------CURRENT LIMITATIONS----------------------------------------------
//
// Only match adjacent instructions in same basic block
// Only equality constraints
// Only constraints between operands, not (0.dest_reg == EAX_enc)
// Only one replacement instruction
//
// ---------EXAMPLE----------------------------------------------------------
//
// // pertinent parts of existing instructions in architecture description
// instruct movI(eRegI dst, eRegI src) %{
// match(Set dst (CopyI src));
// %}
//
// instruct incI_eReg(eRegI dst, immI1 src, eFlagsReg cr) %{
// match(Set dst (AddI dst src));
// effect(KILL cr);
// %}
//
// // Change (inc mov) to lea
// peephole %{
// // increment preceeded by register-register move
// peepmatch ( incI_eReg movI );
// // require that the destination register of the increment
// // match the destination register of the move
// peepconstraint ( 0.dst == 1.dst );
// // construct a replacement instruction that sets
// // the destination to ( move's source register + one )
// peepreplace ( leaI_eReg_immI( 0.dst 1.src 0.src ) );
// %}
//
// Implementation no longer uses movX instructions since
// machine-independent system no longer uses CopyX nodes.
//
// peephole %{
// peepmatch ( incI_eReg movI );
// peepconstraint ( 0.dst == 1.dst );
// peepreplace ( leaI_eReg_immI( 0.dst 1.src 0.src ) );
// %}
//
// peephole %{
// peepmatch ( decI_eReg movI );
// peepconstraint ( 0.dst == 1.dst );
// peepreplace ( leaI_eReg_immI( 0.dst 1.src 0.src ) );
// %}
//
// peephole %{
// peepmatch ( addI_eReg_imm movI );
// peepconstraint ( 0.dst == 1.dst );
// peepreplace ( leaI_eReg_immI( 0.dst 1.src 0.src ) );
// %}
//
// peephole %{
// peepmatch ( addP_eReg_imm movP );
// peepconstraint ( 0.dst == 1.dst );
// peepreplace ( leaP_eReg_immI( 0.dst 1.src 0.src ) );
// %}
// // Change load of spilled value to only a spill
// instruct storeI(memory mem, eRegI src) %{
// match(Set mem (StoreI mem src));
// %}
//
// instruct loadI(eRegI dst, memory mem) %{
// match(Set dst (LoadI mem));
// %}
//
peephole %{
peepmatch ( loadI storeI );
peepconstraint ( 1.src == 0.dst, 1.mem == 0.mem );
peepreplace ( storeI( 1.mem 1.mem 1.src ) );
%}
peephole %{
peepmatch ( loadL storeL );
peepconstraint ( 1.src == 0.dst, 1.mem == 0.mem );
peepreplace ( storeL( 1.mem 1.mem 1.src ) );
%}
peephole %{
peepmatch ( loadP storeP );
peepconstraint ( 1.src == 0.dst, 1.dst == 0.mem );
peepreplace ( storeP( 1.dst 1.dst 1.src ) );
%}
//----------SMARTSPILL RULES---------------------------------------------------
// These must follow all instruction definitions as they use the names
// defined in the instructions definitions.