mirrors/Bochs
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
b6c26d394c
Bochs/bochs/cpu/fetchdecode.cc

2888 lines
133 KiB
C++
Raw Normal View History

- add RCS Id to header of every file. This makes it easier to know what's going on when someone sends in a modified file.
2001-10-03 17:10:38 +04:00
/////////////////////////////////////////////////////////////////////////
a little simpler fetchdecode
2010-05-22 01:17:32 +04:00
// $Id: fetchdecode.cc,v 1.270 2010-05-21 21:17:32 sshwarts Exp $
- add RCS Id to header of every file. This makes it easier to know what's going on when someone sends in a modified file.
2001-10-03 17:10:38 +04:00
/////////////////////////////////////////////////////////////////////////
//
Split shift opcodes | optimize SAR opcode
2010-05-18 11:28:05 +04:00
// Copyright (C) 2001-2010 The Bochs Project
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
//
// This library is free software; you can redistribute it and/or
// modify it under the terms of the GNU Lesser General Public
// License as published by the Free Software Foundation; either
// version 2 of the License, or (at your option) any later version.
//
// This library 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
// Lesser General Public License for more details.
//
// You should have received a copy of the GNU Lesser General Public
// License along with this library; if not, write to the Free Software
- updated FSF address
2009-01-16 21:18:59 +03:00
// Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA B 02110-1301 USA
Cleanups. Move bxInstruction_c definition to separate file instr.h
2008-01-29 20:13:10 +03:00
//
/////////////////////////////////////////////////////////////////////////
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
- parts of the SMP merge apparantly broke the debugger and this revision tries to fix it. The shortcuts to register names such as AX and DL are #defines in cpu/cpu.h, and they are defined in terms of BX_CPU_THIS_PTR. When BX_USE_CPU_SMF=1, this works fine. (This is what bochs used for a long time, and nobody used the SMF=0 mode at all.) To make SMP bochs work, I had to get SMF=0 mode working for the CPU so that there could be an array of cpus. When SMF=0 for the CPU, BX_CPU_THIS_PTR is defined to be "this->" which only works within methods of BX_CPU_C. Code outside of BX_CPU_C must reference BX_CPU(num) instead. - to try to enforce the correct use of AL/AX/DL/etc. shortcuts, they are now only #defined when "NEED_CPU_REG_SHORTCUTS" is #defined. This is only done in the cpu/*.cc code.
2001-05-24 22:46:34 +04:00
#define NEED_CPU_REG_SHORTCUTS 1
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
#include "bochs.h"
Now devices could be compiled separatelly from CPU Averything that required cpu.h include now has it explicitly and there are a lot of files not dependant by CPU at all which will compile a lot faster now ...
2006-03-07 01:03:16 +03:00
#include "cpu.h"
merge in BRANCH-io-cleanup. To see the commit logs for this use either cvsweb or cvs update -r BRANCH-io-cleanup and then 'cvs log' the various files. In general this provides a generic interface for logging. logfunctions:: is a class that is inherited by some classes, and also . allocated as a standalone global called 'genlog'. All logging uses . one of the ::info(), ::error(), ::ldebug(), ::panic() methods of this . class through 'BX_INFO(), BX_ERROR(), BX_DEBUG(), BX_PANIC()' macros . respectively. . . An example usage: . BX_INFO(("Hello, World!\n")); iofunctions:: is a class that is allocated once by default, and assigned as the iofunction of each logfunctions instance. It is this class that maintains the file descriptor and other output related code, at this point using vfprintf(). At some future point, someone may choose to write a gui 'console' for bochs to which messages would be redirected simply by assigning a different iofunction class to the various logfunctions objects. More cleanup is coming, but this works for now. If you want to see alot of debugging output, in main.cc, change onoff[LOGLEV_DEBUG]=0 to =1. Comments, bugs, flames, to me: todd@fries.net
2001-05-15 18:49:57 +04:00
#define LOG_THIS BX_CPU_THIS_PTR
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
///////////////////////////
// prefix bytes
// opcode bytes
// modrm/sib
// address displacement
// immediate constant
///////////////////////////
Fixes in MODRM tables
2008-02-15 15:23:49 +03:00
#define X 0 /* undefined opcode */
Reduce size of Bochs static tables by changing from bx_bool (which is 32bit) to Bit8u
2008-09-06 22:21:29 +04:00
static const Bit8u BxOpcodeHasModrm32[512] = {
Replace BxAnother attribute in fetchdecode by table lookup like it is done in disasm. This is done in preparation to feature huge fetchdecode change - all fethdecode tables will be duplicated and made separatate table for ModC0 and others. So ALL instructions will emjoy SplitMod11b automatically (if they want). After splitting ALL instruction I hope to get 20% speedup at least.
2007-11-15 20:57:56 +03:00
/* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
/* ------------------------------- */
Fixes in MODRM tables
2008-02-15 15:23:49 +03:00
/* 00 */ 1,1,1,1,0,0,0,0,1,1,1,1,0,0,0,X,
white space cleanup
2008-02-03 00:46:54 +03:00
/* 10 */ 1,1,1,1,0,0,0,0,1,1,1,1,0,0,0,0,
Fixes in MODRM tables
2008-02-15 15:23:49 +03:00
/* 20 */ 1,1,1,1,0,0,X,0,1,1,1,1,0,0,X,0,
/* 30 */ 1,1,1,1,0,0,X,0,1,1,1,1,0,0,X,0,
white space cleanup
2008-02-03 00:46:54 +03:00
/* 40 */ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
/* 50 */ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
Fixes in MODRM tables
2008-02-15 15:23:49 +03:00
/* 60 */ 0,0,1,1,X,X,X,X,0,1,0,1,0,0,0,0,
white space cleanup
2008-02-03 00:46:54 +03:00
/* 70 */ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
/* 80 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
/* 90 */ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
/* A0 */ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
/* B0 */ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
/* C0 */ 1,1,0,0,1,1,1,1,0,0,0,0,0,0,0,0,
/* D0 */ 1,1,1,1,0,0,0,0,1,1,1,1,1,1,1,1,
/* E0 */ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
Fixes in MODRM tables
2008-02-15 15:23:49 +03:00
/* F0 */ X,0,X,X,0,0,1,1,0,0,0,0,0,0,1,1,
Replace BxAnother attribute in fetchdecode by table lookup like it is done in disasm. This is done in preparation to feature huge fetchdecode change - all fethdecode tables will be duplicated and made separatate table for ModC0 and others. So ALL instructions will emjoy SplitMod11b automatically (if they want). After splitting ALL instruction I hope to get 20% speedup at least.
2007-11-15 20:57:56 +03:00
/* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
/* ------------------------------- */
Fixes in MODRM tables
2008-02-15 15:23:49 +03:00
1,1,1,1,X,0,0,0,0,0,X,0,X,1,0,1, /* 0F 00 */
Split more opcodes - G3 group
2007-11-17 19:20:37 +03:00
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0F 10 */
Fixes in MODRM tables
2008-02-15 15:23:49 +03:00
1,1,1,1,1,X,1,X,1,1,1,1,1,1,1,1, /* 0F 20 */
0,0,0,0,0,0,X,X,1,X,1,X,X,X,X,X, /* 0F 30 */
Replace BxAnother attribute in fetchdecode by table lookup like it is done in disasm. This is done in preparation to feature huge fetchdecode change - all fethdecode tables will be duplicated and made separatate table for ModC0 and others. So ALL instructions will emjoy SplitMod11b automatically (if they want). After splitting ALL instruction I hope to get 20% speedup at least.
2007-11-15 20:57:56 +03:00
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0F 40 */
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0F 50 */
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0F 60 */
Added Intel VMX emulation to Bochs CPU
2009-01-31 13:43:24 +03:00
1,1,1,1,1,1,1,0,1,1,X,X,1,1,1,1, /* 0F 70 */
Replace BxAnother attribute in fetchdecode by table lookup like it is done in disasm. This is done in preparation to feature huge fetchdecode change - all fethdecode tables will be duplicated and made separatate table for ModC0 and others. So ALL instructions will emjoy SplitMod11b automatically (if they want). After splitting ALL instruction I hope to get 20% speedup at least.
2007-11-15 20:57:56 +03:00
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0F 80 */
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0F 90 */
0,0,0,1,1,1,0,0,0,0,0,1,1,1,1,1, /* 0F A0 */
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0F B0 */
1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0, /* 0F C0 */
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0F D0 */
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0F E0 */
Fixes in MODRM tables
2008-02-15 15:23:49 +03:00
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,X /* 0F F0 */
Replace BxAnother attribute in fetchdecode by table lookup like it is done in disasm. This is done in preparation to feature huge fetchdecode change - all fethdecode tables will be duplicated and made separatate table for ModC0 and others. So ALL instructions will emjoy SplitMod11b automatically (if they want). After splitting ALL instruction I hope to get 20% speedup at least.
2007-11-15 20:57:56 +03:00
/* ------------------------------- */
/* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
};
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
Fixes in MODRM tables
2008-02-15 15:23:49 +03:00
#undef X
Merged patch-unallowed-lock-cases patch. According to the Intel manuals: The LOCK prefix can be prepended only to the following instructions and only to those forms of the instructions where the destination operand is a memory operand: ADD, ADC, AND, BTC, BTR, BTS, CMPXCHG, CMPXCH8B, DEC, INC, NEG, NOT, OR, SBB, SUB, XOR, XADD, and XCHG. If the LOCK prefix is used with one of these instructions and the source operand is a memory operand, an undefined opcode exception (#UD) will be generated. An undefined opcode exception will also be generated if the LOCK prefix is used with any instruction not in the above list. Checking of the LOCK prefix done in fetchDecode state and not overloads Bochs's execution.
2003-04-05 16:16:53 +04:00
/* *********** */
// LOCK PREFIX //
/* *********** */
white space cleanup
2008-02-03 00:46:54 +03:00
/*
Merged patch-unallowed-lock-cases patch. According to the Intel manuals: The LOCK prefix can be prepended only to the following instructions and only to those forms of the instructions where the destination operand is a memory operand: ADD, ADC, AND, BTC, BTR, BTS, CMPXCHG, CMPXCH8B, DEC, INC, NEG, NOT, OR, SBB, SUB, XOR, XADD, and XCHG. If the LOCK prefix is used with one of these instructions and the source operand is a memory operand, an undefined opcode exception (#UD) will be generated. An undefined opcode exception will also be generated if the LOCK prefix is used with any instruction not in the above list. Checking of the LOCK prefix done in fetchDecode state and not overloads Bochs's execution.
2003-04-05 16:16:53 +04:00
* The LOCK prefix can be prepended only to the following instructions
* and only to those forms of the instructions where the destination
* operand is a memory operand: ADD, ADC, AND, BTC, BTR, BTS, CMPXCHG,
* CMPXCH8B, DEC, INC, NEG, NOT, OR, SBB, SUB, XOR, XADD, and XCHG. If
* the LOCK prefix is used with one of these instructions and the source
* operand is a memory operand, an undefined opcode exception (#UD) will
* be generated. An undefined opcode exception will also be generated if
* the LOCK prefix is used with any instruction not in the above list.
* The XCHG instruction always asserts the LOCK# signal regardless of the
* presence or absence of the LOCK prefix.
*/
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
Hide some BX_MEM_C variables Optimize resolve16 methods - by reducing their amount again - reduce chance for misspredictin
2008-05-02 00:08:37 +04:00
static unsigned Resolve16BaseReg[8] = {
BX_16BIT_REG_BX,
BX_16BIT_REG_BX,
BX_16BIT_REG_BP,
BX_16BIT_REG_BP,
BX_16BIT_REG_SI,
BX_16BIT_REG_DI,
BX_16BIT_REG_BP,
BX_16BIT_REG_BX
};
static unsigned Resolve16IndexReg[8] = {
BX_16BIT_REG_SI,
BX_16BIT_REG_DI,
BX_16BIT_REG_SI,
BX_16BIT_REG_DI,
BX_NIL_REGISTER,
BX_NIL_REGISTER,
BX_NIL_REGISTER,
BX_NIL_REGISTER
Fixed CALL/JMP far through call gate 64 Decode SWAPGS and RDTSCP instructions Indent changes in fetchdecode
2006-03-22 23:47:11 +03:00
};
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
make sreg index tables static in fetchdecode and remove them from init.cc/cpu.h
2007-11-19 00:07:40 +03:00
// decoding instructions; accessing seg reg's by index
static unsigned sreg_mod00_rm16[8] = {
BX_SEG_REG_DS,
BX_SEG_REG_DS,
BX_SEG_REG_SS,
BX_SEG_REG_SS,
BX_SEG_REG_DS,
BX_SEG_REG_DS,
BX_SEG_REG_DS,
BX_SEG_REG_DS
};
static unsigned sreg_mod01or10_rm16[8] = {
BX_SEG_REG_DS,
BX_SEG_REG_DS,
BX_SEG_REG_SS,
BX_SEG_REG_SS,
BX_SEG_REG_DS,
BX_SEG_REG_DS,
BX_SEG_REG_SS,
BX_SEG_REG_DS
};
// decoding instructions; accessing seg reg's by index
static unsigned sreg_mod0_base32[8] = {
BX_SEG_REG_DS,
BX_SEG_REG_DS,
BX_SEG_REG_DS,
BX_SEG_REG_DS,
BX_SEG_REG_SS,
BX_SEG_REG_DS,
BX_SEG_REG_DS,
BX_SEG_REG_DS
};
static unsigned sreg_mod1or2_base32[8] = {
BX_SEG_REG_DS,
BX_SEG_REG_DS,
BX_SEG_REG_DS,
BX_SEG_REG_DS,
BX_SEG_REG_SS,
BX_SEG_REG_SS,
BX_SEG_REG_DS,
BX_SEG_REG_DS
};
Merged NEW-INSTRUCTIONS branch
2003-05-15 20:41:17 +04:00
// common fetchdecode32/64 opcode tables
#include "fetchdecode.h"
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
// table of all Bochs opcodes
Clear some duplicated arithmetic opcodes - difference only in operands order
2008-07-13 13:59:59 +04:00
bxIAOpcodeTable BxOpcodesTable[] = {
1st step of moving CPU options to runtime
2010-02-24 22:27:51 +03:00
#define bx_define_opcode(a, b, c, d) { b, c },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
#include "ia_opcodes.h"
};
#undef bx_define_opcode
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* **************************** */
/* 512*2 entries for 16bit mode */
/* 512*2 entries for 32bit mode */
/* **************************** */
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
static const BxOpcodeInfo_t BxOpcodeInfo32[512*2*2] = {
// 512 (/r) + 512 (/m) entries for 16bit mode
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 00 /wr */ { BxArithDstRM, BX_IA_ADD_GbEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 00 /wm */ { BxLockable, BX_IA_ADD_EbGbM },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 01 /wr */ { BxArithDstRM, BX_IA_ADD_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 01 /wm */ { BxLockable, BX_IA_ADD_EwGwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 02 /wr */ { 0, BX_IA_ADD_GbEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 02 /wm */ { 0, BX_IA_ADD_GbEbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 03 /wr */ { 0, BX_IA_ADD_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 03 /wm */ { 0, BX_IA_ADD_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 04 /wr */ { BxImmediate_Ib, BX_IA_ADD_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 04 /wm */ { BxImmediate_Ib, BX_IA_ADD_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 05 /wr */ { BxImmediate_Iw, BX_IA_ADD_AXIw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 05 /wm */ { BxImmediate_Iw, BX_IA_ADD_AXIw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 06 /wr */ { 0, BX_IA_PUSH16_ES },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 06 /wm */ { 0, BX_IA_PUSH16_ES },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 07 /wr */ { 0, BX_IA_POP16_ES },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 07 /wm */ { 0, BX_IA_POP16_ES },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 08 /wr */ { BxArithDstRM, BX_IA_OR_GbEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 08 /wm */ { BxLockable, BX_IA_OR_EbGbM },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 09 /wr */ { BxArithDstRM, BX_IA_OR_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 09 /wm */ { BxLockable, BX_IA_OR_EwGwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0A /wr */ { 0, BX_IA_OR_GbEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0A /wm */ { 0, BX_IA_OR_GbEbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0B /wr */ { 0, BX_IA_OR_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0B /wm */ { 0, BX_IA_OR_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0C /wr */ { BxImmediate_Ib, BX_IA_OR_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0C /wm */ { BxImmediate_Ib, BX_IA_OR_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0D /wr */ { BxImmediate_Iw, BX_IA_OR_AXIw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0D /wm */ { BxImmediate_Iw, BX_IA_OR_AXIw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0E /wr */ { 0, BX_IA_PUSH16_CS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0E /wm */ { 0, BX_IA_PUSH16_CS },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F /wr */ { 0, BX_IA_ERROR }, // 2-byte escape
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F /wm */ { 0, BX_IA_ERROR }, // 2-byte escape
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 10 /wr */ { BxArithDstRM, BX_IA_ADC_GbEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 10 /wm */ { BxLockable, BX_IA_ADC_EbGbM },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 11 /wr */ { BxArithDstRM, BX_IA_ADC_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 11 /wm */ { BxLockable, BX_IA_ADC_EwGwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 12 /wr */ { 0, BX_IA_ADC_GbEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 12 /wm */ { 0, BX_IA_ADC_GbEbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 13 /wr */ { 0, BX_IA_ADC_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 13 /wm */ { 0, BX_IA_ADC_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 14 /wr */ { BxImmediate_Ib, BX_IA_ADC_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 14 /wm */ { BxImmediate_Ib, BX_IA_ADC_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 15 /wr */ { BxImmediate_Iw, BX_IA_ADC_AXIw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 15 /wm */ { BxImmediate_Iw, BX_IA_ADC_AXIw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 16 /wr */ { 0, BX_IA_PUSH16_SS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 16 /wm */ { 0, BX_IA_PUSH16_SS },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 17 /wr */ { BxTraceEnd, BX_IA_POP16_SS }, // async_event = 1
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 17 /wm */ { BxTraceEnd, BX_IA_POP16_SS }, // async_event = 1
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 18 /wr */ { BxArithDstRM, BX_IA_SBB_GbEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 18 /wm */ { BxLockable, BX_IA_SBB_EbGbM },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 19 /wr */ { BxArithDstRM, BX_IA_SBB_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 19 /wm */ { BxLockable, BX_IA_SBB_EwGwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 1A /wr */ { 0, BX_IA_SBB_GbEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 1A /wm */ { 0, BX_IA_SBB_GbEbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 1B /wr */ { 0, BX_IA_SBB_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 1B /wm */ { 0, BX_IA_SBB_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 1C /wr */ { BxImmediate_Ib, BX_IA_SBB_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 1C /wm */ { BxImmediate_Ib, BX_IA_SBB_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 1D /wr */ { BxImmediate_Iw, BX_IA_SBB_AXIw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 1D /wm */ { BxImmediate_Iw, BX_IA_SBB_AXIw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 1E /wr */ { 0, BX_IA_PUSH16_DS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 1E /wm */ { 0, BX_IA_PUSH16_DS },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 1F /wr */ { 0, BX_IA_POP16_DS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 1F /wm */ { 0, BX_IA_POP16_DS },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 20 /wr */ { BxArithDstRM, BX_IA_AND_GbEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 20 /wm */ { BxLockable, BX_IA_AND_EbGbM },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 21 /wr */ { BxArithDstRM, BX_IA_AND_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 21 /wm */ { BxLockable, BX_IA_AND_EwGwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 22 /wr */ { 0, BX_IA_AND_GbEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 22 /wm */ { 0, BX_IA_AND_GbEbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 23 /wr */ { 0, BX_IA_AND_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 23 /wm */ { 0, BX_IA_AND_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 24 /wr */ { BxImmediate_Ib, BX_IA_AND_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 24 /wm */ { BxImmediate_Ib, BX_IA_AND_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 25 /wr */ { BxImmediate_Iw, BX_IA_AND_AXIw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 25 /wm */ { BxImmediate_Iw, BX_IA_AND_AXIw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 26 /wr */ { 0, BX_IA_ERROR }, // ES:
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 26 /wm */ { 0, BX_IA_ERROR }, // ES:
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 27 /wr */ { 0, BX_IA_DAA },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 27 /wm */ { 0, BX_IA_DAA },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 28 /wr */ { BxArithDstRM, BX_IA_SUB_GbEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 28 /wm */ { BxLockable, BX_IA_SUB_EbGbM },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 29 /wr */ { BxArithDstRM, BX_IA_SUB_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 29 /wm */ { BxLockable, BX_IA_SUB_EwGwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 2A /wr */ { 0, BX_IA_SUB_GbEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 2A /wm */ { 0, BX_IA_SUB_GbEbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 2B /wr */ { 0, BX_IA_SUB_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 2B /wm */ { 0, BX_IA_SUB_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 2C /wr */ { BxImmediate_Ib, BX_IA_SUB_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 2C /wm */ { BxImmediate_Ib, BX_IA_SUB_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 2D /wr */ { BxImmediate_Iw, BX_IA_SUB_AXIw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 2D /wm */ { BxImmediate_Iw, BX_IA_SUB_AXIw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 2E /wr */ { 0, BX_IA_ERROR }, // CS:
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 2E /wm */ { 0, BX_IA_ERROR }, // CS:
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 2F /wr */ { 0, BX_IA_DAS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 2F /wm */ { 0, BX_IA_DAS },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 30 /wr */ { BxArithDstRM, BX_IA_XOR_GbEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 30 /wm */ { BxLockable, BX_IA_XOR_EbGbM },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 31 /wr */ { BxArithDstRM, BX_IA_XOR_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 31 /wm */ { BxLockable, BX_IA_XOR_EwGwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 32 /wr */ { 0, BX_IA_XOR_GbEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 32 /wm */ { 0, BX_IA_XOR_GbEbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 33 /wr */ { 0, BX_IA_XOR_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 33 /wm */ { 0, BX_IA_XOR_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 34 /wr */ { BxImmediate_Ib, BX_IA_XOR_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 34 /wm */ { BxImmediate_Ib, BX_IA_XOR_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 35 /wr */ { BxImmediate_Iw, BX_IA_XOR_AXIw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 35 /wm */ { BxImmediate_Iw, BX_IA_XOR_AXIw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 36 /wr */ { 0, BX_IA_ERROR }, // SS:
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 36 /wm */ { 0, BX_IA_ERROR }, // SS:
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 37 /wr */ { 0, BX_IA_AAA },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 37 /wm */ { 0, BX_IA_AAA },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 38 /wr */ { BxArithDstRM, BX_IA_CMP_GbEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 38 /wm */ { 0, BX_IA_CMP_EbGbM },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 39 /wr */ { BxArithDstRM, BX_IA_CMP_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 39 /wm */ { 0, BX_IA_CMP_EwGwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 3A /wr */ { 0, BX_IA_CMP_GbEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 3A /wm */ { 0, BX_IA_CMP_GbEbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 3B /wr */ { 0, BX_IA_CMP_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 3B /wm */ { 0, BX_IA_CMP_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 3C /wr */ { BxImmediate_Ib, BX_IA_CMP_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 3C /wm */ { BxImmediate_Ib, BX_IA_CMP_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 3D /wr */ { BxImmediate_Iw, BX_IA_CMP_AXIw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 3D /wm */ { BxImmediate_Iw, BX_IA_CMP_AXIw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 3E /wr */ { 0, BX_IA_ERROR }, // DS:
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 3E /wm */ { 0, BX_IA_ERROR }, // DS:
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 3F /wr */ { 0, BX_IA_AAS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 3F /wm */ { 0, BX_IA_AAS },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 40 /wr */ { 0, BX_IA_INC_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 40 /wm */ { 0, BX_IA_INC_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 41 /wr */ { 0, BX_IA_INC_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 41 /wm */ { 0, BX_IA_INC_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 42 /wr */ { 0, BX_IA_INC_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 42 /wm */ { 0, BX_IA_INC_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 43 /wr */ { 0, BX_IA_INC_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 43 /wm */ { 0, BX_IA_INC_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 44 /wr */ { 0, BX_IA_INC_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 44 /wm */ { 0, BX_IA_INC_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 45 /wr */ { 0, BX_IA_INC_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 45 /wm */ { 0, BX_IA_INC_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 46 /wr */ { 0, BX_IA_INC_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 46 /wm */ { 0, BX_IA_INC_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 47 /wr */ { 0, BX_IA_INC_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 47 /wm */ { 0, BX_IA_INC_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 48 /wr */ { 0, BX_IA_DEC_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 48 /wm */ { 0, BX_IA_DEC_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 49 /wr */ { 0, BX_IA_DEC_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 49 /wm */ { 0, BX_IA_DEC_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 4A /wr */ { 0, BX_IA_DEC_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 4A /wm */ { 0, BX_IA_DEC_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 4B /wr */ { 0, BX_IA_DEC_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 4B /wm */ { 0, BX_IA_DEC_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 4C /wr */ { 0, BX_IA_DEC_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 4C /wm */ { 0, BX_IA_DEC_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 4D /wr */ { 0, BX_IA_DEC_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 4D /wm */ { 0, BX_IA_DEC_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 4E /wr */ { 0, BX_IA_DEC_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 4E /wm */ { 0, BX_IA_DEC_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 4F /wr */ { 0, BX_IA_DEC_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 4F /wm */ { 0, BX_IA_DEC_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 50 /wr */ { 0, BX_IA_PUSH_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 50 /wm */ { 0, BX_IA_PUSH_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 51 /wr */ { 0, BX_IA_PUSH_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 51 /wm */ { 0, BX_IA_PUSH_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 52 /wr */ { 0, BX_IA_PUSH_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 52 /wm */ { 0, BX_IA_PUSH_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 53 /wr */ { 0, BX_IA_PUSH_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 53 /wm */ { 0, BX_IA_PUSH_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 54 /wr */ { 0, BX_IA_PUSH_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 54 /wm */ { 0, BX_IA_PUSH_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 55 /wr */ { 0, BX_IA_PUSH_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 55 /wm */ { 0, BX_IA_PUSH_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 56 /wr */ { 0, BX_IA_PUSH_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 56 /wm */ { 0, BX_IA_PUSH_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 57 /wr */ { 0, BX_IA_PUSH_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 57 /wm */ { 0, BX_IA_PUSH_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 58 /wr */ { 0, BX_IA_POP_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 58 /wm */ { 0, BX_IA_POP_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 59 /wr */ { 0, BX_IA_POP_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 59 /wm */ { 0, BX_IA_POP_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 5A /wr */ { 0, BX_IA_POP_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 5A /wm */ { 0, BX_IA_POP_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 5B /wr */ { 0, BX_IA_POP_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 5B /wm */ { 0, BX_IA_POP_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 5C /wr */ { 0, BX_IA_POP_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 5C /wm */ { 0, BX_IA_POP_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 5D /wr */ { 0, BX_IA_POP_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 5D /wm */ { 0, BX_IA_POP_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 5E /wr */ { 0, BX_IA_POP_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 5E /wm */ { 0, BX_IA_POP_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 5F /wr */ { 0, BX_IA_POP_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 5F /wm */ { 0, BX_IA_POP_RX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 60 /wr */ { 0, BX_IA_PUSHAD16 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 60 /wm */ { 0, BX_IA_PUSHAD16 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 61 /wr */ { 0, BX_IA_POPAD16 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 61 /wm */ { 0, BX_IA_POPAD16 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 62 /wr */ { 0, BX_IA_ERROR }, // BOUND
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 62 /wm */ { 0, BX_IA_BOUND_GwMa },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 63 /wr */ { 0, BX_IA_ARPL_EwGw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 63 /wm */ { 0, BX_IA_ARPL_EwGw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 64 /wr */ { 0, BX_IA_ERROR }, // FS:
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 64 /wm */ { 0, BX_IA_ERROR }, // FS:
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 65 /wr */ { 0, BX_IA_ERROR }, // GS:
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 65 /wm */ { 0, BX_IA_ERROR }, // GS:
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 66 /wr */ { 0, BX_IA_ERROR }, // OS:
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 66 /wm */ { 0, BX_IA_ERROR }, // OS:
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 67 /wr */ { 0, BX_IA_ERROR }, // AS:
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 67 /wm */ { 0, BX_IA_ERROR }, // AS:
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 68 /wr */ { BxImmediate_Iw, BX_IA_PUSH_Iw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 68 /wm */ { BxImmediate_Iw, BX_IA_PUSH_Iw },
Split more opcodes using new LOAD technique
2008-08-11 01:16:12 +04:00
/* 69 /wr */ { BxImmediate_Iw, BX_IA_IMUL_GwEwIwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 69 /wm */ { BxImmediate_Iw, BX_IA_IMUL_GwEwIwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 6A /wr */ { BxImmediate_Ib_SE, BX_IA_PUSH_Iw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 6A /wm */ { BxImmediate_Ib_SE, BX_IA_PUSH_Iw },
Split more opcodes using new LOAD technique
2008-08-11 01:16:12 +04:00
/* 6B /wr */ { BxImmediate_Ib_SE, BX_IA_IMUL_GwEwIwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 6B /wm */ { BxImmediate_Ib_SE, BX_IA_IMUL_GwEwIwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 6C /wr */ { 0, BX_IA_REP_INSB_YbDX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 6C /wm */ { 0, BX_IA_REP_INSB_YbDX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 6D /wr */ { 0, BX_IA_REP_INSW_YwDX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 6D /wm */ { 0, BX_IA_REP_INSW_YwDX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 6E /wr */ { 0, BX_IA_REP_OUTSB_DXXb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 6E /wm */ { 0, BX_IA_REP_OUTSB_DXXb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 6F /wr */ { 0, BX_IA_REP_OUTSW_DXXw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 6F /wm */ { 0, BX_IA_REP_OUTSW_DXXw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 70 /wr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JO_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 70 /wm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JO_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 71 /wr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNO_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 71 /wm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNO_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 72 /wr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JB_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 72 /wm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JB_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 73 /wr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNB_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 73 /wm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNB_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 74 /wr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JZ_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 74 /wm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JZ_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 75 /wr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNZ_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 75 /wm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNZ_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 76 /wr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JBE_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 76 /wm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JBE_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 77 /wr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNBE_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 77 /wm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNBE_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 78 /wr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JS_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 78 /wm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JS_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 79 /wr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNS_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 79 /wm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNS_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 7A /wr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JP_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 7A /wm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JP_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 7B /wr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNP_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 7B /wm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNP_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 7C /wr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JL_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 7C /wm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JL_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 7D /wr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNL_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 7D /wm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNL_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 7E /wr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JLE_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 7E /wm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JLE_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 7F /wr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNLE_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 7F /wm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNLE_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 80 /wr */ { BxGroup1 | BxImmediate_Ib, BX_IA_ERROR, BxOpcodeInfoG1EbIbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 80 /wm */ { BxGroup1 | BxImmediate_Ib, BX_IA_ERROR, BxOpcodeInfoG1EbIbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 81 /wr */ { BxGroup1 | BxImmediate_Iw, BX_IA_ERROR, BxOpcodeInfoG1EwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 81 /wm */ { BxGroup1 | BxImmediate_Iw, BX_IA_ERROR, BxOpcodeInfoG1EwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 82 /wr */ { BxGroup1 | BxImmediate_Ib, BX_IA_ERROR, BxOpcodeInfoG1EbIbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 82 /wm */ { BxGroup1 | BxImmediate_Ib, BX_IA_ERROR, BxOpcodeInfoG1EbIbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 83 /wr */ { BxGroup1 | BxImmediate_Ib_SE, BX_IA_ERROR, BxOpcodeInfoG1EwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 83 /wm */ { BxGroup1 | BxImmediate_Ib_SE, BX_IA_ERROR, BxOpcodeInfoG1EwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 84 /wr */ { 0, BX_IA_TEST_EbGbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 84 /wm */ { 0, BX_IA_TEST_EbGbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 85 /wr */ { 0, BX_IA_TEST_EwGwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 85 /wm */ { 0, BX_IA_TEST_EwGwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 86 /wr */ { 0, BX_IA_XCHG_EbGbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 86 /wm */ { BxLockable, BX_IA_XCHG_EbGbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 87 /wr */ { 0, BX_IA_XCHG_EwGwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 87 /wm */ { BxLockable, BX_IA_XCHG_EwGwM },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 88 /wr */ { BxArithDstRM, BX_IA_MOV_GbEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 88 /wm */ { 0, BX_IA_MOV_EbGbM },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 89 /wr */ { BxArithDstRM, BX_IA_MOV_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 89 /wm */ { 0, BX_IA_MOV_EwGwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 8A /wr */ { 0, BX_IA_MOV_GbEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 8A /wm */ { 0, BX_IA_MOV_GbEbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 8B /wr */ { 0, BX_IA_MOV_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 8B /wm */ { 0, BX_IA_MOV_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 8C /wr */ { 0, BX_IA_MOV_EwSwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 8C /wm */ { 0, BX_IA_MOV_EwSwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 8D /wr */ { 0, BX_IA_ERROR }, // LEA
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 8D /wm */ { 0, BX_IA_LEA_GwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 8E /wr */ { 0, BX_IA_MOV_SwEw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 8E /wm */ { 0, BX_IA_MOV_SwEw },
More fetchdecode fixes
2008-05-01 01:32:33 +04:00
/* 8F /wr */ { BxGroup1A, BX_IA_ERROR, BxOpcodeInfoG1AEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 8F /wm */ { BxGroup1A, BX_IA_ERROR, BxOpcodeInfoG1AEwM },
Separate PAUSE instruction from regular NOP
2009-01-27 23:29:05 +03:00
/* 90 /wr */ { BxPrefixSSE, BX_IA_NOP, BxOpcodeGroupSSE_PAUSE },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 90 /wm */ { BxPrefixSSE, BX_IA_NOP, BxOpcodeGroupSSE_PAUSE },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 91 /wr */ { 0, BX_IA_XCHG_RXAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 91 /wm */ { 0, BX_IA_XCHG_RXAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 92 /wr */ { 0, BX_IA_XCHG_RXAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 92 /wm */ { 0, BX_IA_XCHG_RXAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 93 /wr */ { 0, BX_IA_XCHG_RXAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 93 /wm */ { 0, BX_IA_XCHG_RXAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 94 /wr */ { 0, BX_IA_XCHG_RXAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 94 /wm */ { 0, BX_IA_XCHG_RXAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 95 /wr */ { 0, BX_IA_XCHG_RXAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 95 /wm */ { 0, BX_IA_XCHG_RXAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 96 /wr */ { 0, BX_IA_XCHG_RXAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 96 /wm */ { 0, BX_IA_XCHG_RXAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 97 /wr */ { 0, BX_IA_XCHG_RXAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 97 /wm */ { 0, BX_IA_XCHG_RXAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 98 /wr */ { 0, BX_IA_CBW },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 98 /wm */ { 0, BX_IA_CBW },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 99 /wr */ { 0, BX_IA_CWD },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 99 /wm */ { 0, BX_IA_CWD },
branch_16 optimizations
2008-08-24 02:27:58 +04:00
/* 9A /wr */ { BxImmediate_IwIw | BxTraceEnd, BX_IA_CALL16_Ap },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 9A /wm */ { BxImmediate_IwIw | BxTraceEnd, BX_IA_CALL16_Ap },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 9B /wr */ { 0, BX_IA_FWAIT },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 9B /wm */ { 0, BX_IA_FWAIT },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 9C /wr */ { 0, BX_IA_PUSHF_Fw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 9C /wm */ { 0, BX_IA_PUSHF_Fw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 9D /wr */ { 0, BX_IA_POPF_Fw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 9D /wm */ { 0, BX_IA_POPF_Fw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 9E /wr */ { 0, BX_IA_SAHF },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 9E /wm */ { 0, BX_IA_SAHF },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 9F /wr */ { 0, BX_IA_LAHF },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 9F /wm */ { 0, BX_IA_LAHF },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A0 /wr */ { BxImmediate_O, BX_IA_MOV_ALOd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A0 /wm */ { BxImmediate_O, BX_IA_MOV_ALOd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A1 /wr */ { BxImmediate_O, BX_IA_MOV_AXOd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A1 /wm */ { BxImmediate_O, BX_IA_MOV_AXOd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A2 /wr */ { BxImmediate_O, BX_IA_MOV_OdAL },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A2 /wm */ { BxImmediate_O, BX_IA_MOV_OdAL },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A3 /wr */ { BxImmediate_O, BX_IA_MOV_OdAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A3 /wm */ { BxImmediate_O, BX_IA_MOV_OdAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A4 /wr */ { 0, BX_IA_REP_MOVSB_XbYb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A4 /wm */ { 0, BX_IA_REP_MOVSB_XbYb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A5 /wr */ { 0, BX_IA_REP_MOVSW_XwYw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A5 /wm */ { 0, BX_IA_REP_MOVSW_XwYw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A6 /wr */ { 0, BX_IA_REP_CMPSB_XbYb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A6 /wm */ { 0, BX_IA_REP_CMPSB_XbYb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A7 /wr */ { 0, BX_IA_REP_CMPSW_XwYw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A7 /wm */ { 0, BX_IA_REP_CMPSW_XwYw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A8 /wr */ { BxImmediate_Ib, BX_IA_TEST_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A8 /wm */ { BxImmediate_Ib, BX_IA_TEST_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A9 /wr */ { BxImmediate_Iw, BX_IA_TEST_AXIw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A9 /wm */ { BxImmediate_Iw, BX_IA_TEST_AXIw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* AA /wr */ { 0, BX_IA_REP_STOSB_YbAL },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* AA /wm */ { 0, BX_IA_REP_STOSB_YbAL },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* AB /wr */ { 0, BX_IA_REP_STOSW_YwAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* AB /wm */ { 0, BX_IA_REP_STOSW_YwAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* AC /wr */ { 0, BX_IA_REP_LODSB_ALXb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* AC /wm */ { 0, BX_IA_REP_LODSB_ALXb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* AD /wr */ { 0, BX_IA_REP_LODSW_AXXw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* AD /wm */ { 0, BX_IA_REP_LODSW_AXXw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* AE /wr */ { 0, BX_IA_REP_SCASB_ALXb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* AE /wm */ { 0, BX_IA_REP_SCASB_ALXb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* AF /wr */ { 0, BX_IA_REP_SCASW_AXXw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* AF /wm */ { 0, BX_IA_REP_SCASW_AXXw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B0 /wr */ { BxImmediate_Ib, BX_IA_MOV_RLIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B0 /wm */ { BxImmediate_Ib, BX_IA_MOV_RLIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B1 /wr */ { BxImmediate_Ib, BX_IA_MOV_RLIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B1 /wm */ { BxImmediate_Ib, BX_IA_MOV_RLIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B2 /wr */ { BxImmediate_Ib, BX_IA_MOV_RLIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B2 /wm */ { BxImmediate_Ib, BX_IA_MOV_RLIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B3 /wr */ { BxImmediate_Ib, BX_IA_MOV_RLIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B3 /wm */ { BxImmediate_Ib, BX_IA_MOV_RLIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B4 /wr */ { BxImmediate_Ib, BX_IA_MOV_RHIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B4 /wm */ { BxImmediate_Ib, BX_IA_MOV_RHIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B5 /wr */ { BxImmediate_Ib, BX_IA_MOV_RHIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B5 /wm */ { BxImmediate_Ib, BX_IA_MOV_RHIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B6 /wr */ { BxImmediate_Ib, BX_IA_MOV_RHIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B6 /wm */ { BxImmediate_Ib, BX_IA_MOV_RHIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B7 /wr */ { BxImmediate_Ib, BX_IA_MOV_RHIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B7 /wm */ { BxImmediate_Ib, BX_IA_MOV_RHIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B8 /wr */ { BxImmediate_Iw, BX_IA_MOV_RXIw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B8 /wm */ { BxImmediate_Iw, BX_IA_MOV_RXIw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B9 /wr */ { BxImmediate_Iw, BX_IA_MOV_RXIw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B9 /wm */ { BxImmediate_Iw, BX_IA_MOV_RXIw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* BA /wr */ { BxImmediate_Iw, BX_IA_MOV_RXIw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* BA /wm */ { BxImmediate_Iw, BX_IA_MOV_RXIw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* BB /wr */ { BxImmediate_Iw, BX_IA_MOV_RXIw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* BB /wm */ { BxImmediate_Iw, BX_IA_MOV_RXIw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* BC /wr */ { BxImmediate_Iw, BX_IA_MOV_RXIw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* BC /wm */ { BxImmediate_Iw, BX_IA_MOV_RXIw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* BD /wr */ { BxImmediate_Iw, BX_IA_MOV_RXIw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* BD /wm */ { BxImmediate_Iw, BX_IA_MOV_RXIw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* BE /wr */ { BxImmediate_Iw, BX_IA_MOV_RXIw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* BE /wm */ { BxImmediate_Iw, BX_IA_MOV_RXIw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* BF /wr */ { BxImmediate_Iw, BX_IA_MOV_RXIw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* BF /wm */ { BxImmediate_Iw, BX_IA_MOV_RXIw },
Split shift opcodes | optimize SAR opcode
2010-05-18 11:28:05 +04:00
/* C0 /wr */ { BxGroup2 | BxImmediate_Ib, BX_IA_ERROR, BxOpcodeInfoG2EbR },
/* C0 /wm */ { BxGroup2 | BxImmediate_Ib, BX_IA_ERROR, BxOpcodeInfoG2EbM },
/* C1 /wr */ { BxGroup2 | BxImmediate_Ib, BX_IA_ERROR, BxOpcodeInfoG2EwR },
/* C1 /wm */ { BxGroup2 | BxImmediate_Ib, BX_IA_ERROR, BxOpcodeInfoG2EwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* C2 /wr */ { BxImmediate_Iw | BxTraceEnd, BX_IA_RETnear16_Iw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* C2 /wm */ { BxImmediate_Iw | BxTraceEnd, BX_IA_RETnear16_Iw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* C3 /wr */ { BxTraceEnd, BX_IA_RETnear16 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* C3 /wm */ { BxTraceEnd, BX_IA_RETnear16 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* C4 /wr */ { 0, BX_IA_ERROR }, // LES
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* C4 /wm */ { 0, BX_IA_LES_GwMp },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* C5 /wr */ { 0, BX_IA_ERROR }, // LDS
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* C5 /wm */ { 0, BX_IA_LDS_GwMp },
Correctly implement opcode group G11 i.e. instructions C6 and C7 should @UD when modrm nnn field != 0 (1st instr in the group
2008-04-25 01:52:28 +04:00
/* C6 /wr */ { BxGroup11, BX_IA_ERROR, BxOpcodeInfoG11EbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* C6 /wm */ { BxGroup11, BX_IA_ERROR, BxOpcodeInfoG11EbM },
Correctly implement opcode group G11 i.e. instructions C6 and C7 should @UD when modrm nnn field != 0 (1st instr in the group
2008-04-25 01:52:28 +04:00
/* C7 /wr */ { BxGroup11, BX_IA_ERROR, BxOpcodeInfoG11EwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* C7 /wm */ { BxGroup11, BX_IA_ERROR, BxOpcodeInfoG11EwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* C8 /wr */ { BxImmediate_IwIb, BX_IA_ENTER16_IwIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* C8 /wm */ { BxImmediate_IwIb, BX_IA_ENTER16_IwIb },
optimize LEAVE operation
2008-08-28 01:57:40 +04:00
/* C9 /wr */ { 0, BX_IA_LEAVE16 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* C9 /wm */ { 0, BX_IA_LEAVE16 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* CA /wr */ { BxImmediate_Iw | BxTraceEnd, BX_IA_RETfar16_Iw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* CA /wm */ { BxImmediate_Iw | BxTraceEnd, BX_IA_RETfar16_Iw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* CB /wr */ { BxTraceEnd, BX_IA_RETfar16 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* CB /wm */ { BxTraceEnd, BX_IA_RETfar16 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* CC /wr */ { BxTraceEnd, BX_IA_INT3 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* CC /wm */ { BxTraceEnd, BX_IA_INT3 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* CD /wr */ { BxImmediate_Ib | BxTraceEnd, BX_IA_INT_Ib },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* CD /wm */ { BxImmediate_Ib | BxTraceEnd, BX_IA_INT_Ib },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* CE /wr */ { BxTraceEnd, BX_IA_INTO },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* CE /wm */ { BxTraceEnd, BX_IA_INTO },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* CF /wr */ { BxTraceEnd, BX_IA_IRET16 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* CF /wm */ { BxTraceEnd, BX_IA_IRET16 },
Split shift opcodes | optimize SAR opcode
2010-05-18 11:28:05 +04:00
/* D0 /wr */ { BxGroup2 | BxImmediate_I1, BX_IA_ERROR, BxOpcodeInfoG2EbR },
/* D0 /wm */ { BxGroup2 | BxImmediate_I1, BX_IA_ERROR, BxOpcodeInfoG2EbM },
/* D1 /wr */ { BxGroup2 | BxImmediate_I1, BX_IA_ERROR, BxOpcodeInfoG2EwR },
/* D1 /wm */ { BxGroup2 | BxImmediate_I1, BX_IA_ERROR, BxOpcodeInfoG2EwM },
/* D2 /wr */ { BxGroup2, BX_IA_ERROR, BxOpcodeInfoG2EbR },
/* D2 /wm */ { BxGroup2, BX_IA_ERROR, BxOpcodeInfoG2EbM },
/* D3 /wr */ { BxGroup2, BX_IA_ERROR, BxOpcodeInfoG2EwR },
/* D3 /wm */ { BxGroup2, BX_IA_ERROR, BxOpcodeInfoG2EwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* D4 /wr */ { BxImmediate_Ib, BX_IA_AAM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* D4 /wm */ { BxImmediate_Ib, BX_IA_AAM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* D5 /wr */ { BxImmediate_Ib, BX_IA_AAD },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* D5 /wm */ { BxImmediate_Ib, BX_IA_AAD },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* D6 /wr */ { 0, BX_IA_SALC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* D6 /wm */ { 0, BX_IA_SALC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* D7 /wr */ { 0, BX_IA_XLAT },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* D7 /wm */ { 0, BX_IA_XLAT },
port some changes from spftfloat-fpu branch to the MT
2004-04-09 19:34:59 +04:00
#if BX_SUPPORT_FPU
optimize some of x87 tables
2010-04-16 23:52:44 +04:00
/* D8 /wr */ { BxGroupN, BX_IA_ERROR, BxOpcodeInfo_FloatingPointD8 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* D8 /wm */ { BxGroupFP, BX_IA_ERROR, BxOpcodeInfo_FPGroupD8 },
A bit simplify FPU decoding
2008-09-16 22:28:53 +04:00
/* D9 /wr */ { BxFPEscape, BX_IA_ERROR, BxOpcodeInfo_FloatingPointD9 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* D9 /wm */ { BxGroupFP, BX_IA_ERROR, BxOpcodeInfo_FPGroupD9 },
A bit simplify FPU decoding
2008-09-16 22:28:53 +04:00
/* DA /wr */ { BxFPEscape, BX_IA_ERROR, BxOpcodeInfo_FloatingPointDA },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DA /wm */ { BxGroupFP, BX_IA_ERROR, BxOpcodeInfo_FPGroupDA },
A bit simplify FPU decoding
2008-09-16 22:28:53 +04:00
/* DB /wr */ { BxFPEscape, BX_IA_ERROR, BxOpcodeInfo_FloatingPointDB },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DB /wm */ { BxGroupFP, BX_IA_ERROR, BxOpcodeInfo_FPGroupDB },
optimize some of x87 tables
2010-04-16 23:52:44 +04:00
/* DC /wr */ { BxGroupN, BX_IA_ERROR, BxOpcodeInfo_FloatingPointDC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DC /wm */ { BxGroupFP, BX_IA_ERROR, BxOpcodeInfo_FPGroupDC },
optimize some of x87 tables
2010-04-16 23:52:44 +04:00
/* DD /wr */ { BxGroupN, BX_IA_ERROR, BxOpcodeInfo_FloatingPointDD },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DD /wm */ { BxGroupFP, BX_IA_ERROR, BxOpcodeInfo_FPGroupDD },
A bit simplify FPU decoding
2008-09-16 22:28:53 +04:00
/* DE /wr */ { BxFPEscape, BX_IA_ERROR, BxOpcodeInfo_FloatingPointDE },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DE /wm */ { BxGroupFP, BX_IA_ERROR, BxOpcodeInfo_FPGroupDE },
A bit simplify FPU decoding
2008-09-16 22:28:53 +04:00
/* DF /wr */ { BxFPEscape, BX_IA_ERROR, BxOpcodeInfo_FloatingPointDF },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DF /wm */ { BxGroupFP, BX_IA_ERROR, BxOpcodeInfo_FPGroupDF },
port some changes from spftfloat-fpu branch to the MT
2004-04-09 19:34:59 +04:00
#else
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* D8 /wr */ { 0, BX_IA_FPU_ESC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* D8 /wm */ { 0, BX_IA_FPU_ESC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* D9 /wr */ { 0, BX_IA_FPU_ESC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* D9 /wm */ { 0, BX_IA_FPU_ESC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* DA /wr */ { 0, BX_IA_FPU_ESC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DA /wm */ { 0, BX_IA_FPU_ESC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* DB /wr */ { 0, BX_IA_FPU_ESC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DB /wm */ { 0, BX_IA_FPU_ESC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* DC /wr */ { 0, BX_IA_FPU_ESC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DC /wm */ { 0, BX_IA_FPU_ESC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* DD /wr */ { 0, BX_IA_FPU_ESC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DD /wm */ { 0, BX_IA_FPU_ESC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* DE /wr */ { 0, BX_IA_FPU_ESC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DE /wm */ { 0, BX_IA_FPU_ESC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* DF /wr */ { 0, BX_IA_FPU_ESC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DF /wm */ { 0, BX_IA_FPU_ESC },
port some changes from spftfloat-fpu branch to the MT
2004-04-09 19:34:59 +04:00
#endif
Reorganize ctrl_xfer8.cc code, allows to inline branch32 method
2008-06-22 07:45:55 +04:00
/* E0 /wr */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_LOOPNE16_Jb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E0 /wm */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_LOOPNE16_Jb },
Reorganize ctrl_xfer8.cc code, allows to inline branch32 method
2008-06-22 07:45:55 +04:00
/* E1 /wr */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_LOOPE16_Jb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E1 /wm */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_LOOPE16_Jb },
Reorganize ctrl_xfer8.cc code, allows to inline branch32 method
2008-06-22 07:45:55 +04:00
/* E2 /wr */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_LOOP16_Jb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E2 /wm */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_LOOP16_Jb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* E3 /wr */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_JCXZ_Jb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E3 /wm */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_JCXZ_Jb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* E4 /wr */ { BxImmediate_Ib, BX_IA_IN_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E4 /wm */ { BxImmediate_Ib, BX_IA_IN_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* E5 /wr */ { BxImmediate_Ib, BX_IA_IN_AXIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E5 /wm */ { BxImmediate_Ib, BX_IA_IN_AXIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* E6 /wr */ { BxImmediate_Ib, BX_IA_OUT_IbAL },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E6 /wm */ { BxImmediate_Ib, BX_IA_OUT_IbAL },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* E7 /wr */ { BxImmediate_Ib, BX_IA_OUT_IbAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E7 /wm */ { BxImmediate_Ib, BX_IA_OUT_IbAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* E8 /wr */ { BxImmediate_BrOff16 | BxTraceEnd, BX_IA_CALL_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E8 /wm */ { BxImmediate_BrOff16 | BxTraceEnd, BX_IA_CALL_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* E9 /wr */ { BxImmediate_BrOff16 | BxTraceEnd, BX_IA_JMP_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E9 /wm */ { BxImmediate_BrOff16 | BxTraceEnd, BX_IA_JMP_Jw },
branch_16 optimizations
2008-08-24 02:27:58 +04:00
/* EA /wr */ { BxImmediate_IwIw | BxTraceEnd, BX_IA_JMP_Ap },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* EA /wm */ { BxImmediate_IwIw | BxTraceEnd, BX_IA_JMP_Ap },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* EB /wr */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_JMP_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* EB /wm */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_JMP_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* EC /wr */ { 0, BX_IA_IN_ALDX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* EC /wm */ { 0, BX_IA_IN_ALDX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* ED /wr */ { 0, BX_IA_IN_AXDX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* ED /wm */ { 0, BX_IA_IN_AXDX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* EE /wr */ { 0, BX_IA_OUT_DXAL },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* EE /wm */ { 0, BX_IA_OUT_DXAL },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* EF /wr */ { 0, BX_IA_OUT_DXAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* EF /wm */ { 0, BX_IA_OUT_DXAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F0 /wr */ { 0, BX_IA_ERROR }, // LOCK
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F0 /wm */ { 0, BX_IA_ERROR }, // LOCK
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F1 /wr */ { BxTraceEnd, BX_IA_INT1 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F1 /wm */ { BxTraceEnd, BX_IA_INT1 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F2 /wr */ { 0, BX_IA_ERROR }, // REPNE/REPNZ
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F2 /wm */ { 0, BX_IA_ERROR }, // REPNE/REPNZ
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F3 /wr */ { 0, BX_IA_ERROR }, // REP, REPE/REPZ
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F3 /wm */ { 0, BX_IA_ERROR }, // REP, REPE/REPZ
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F4 /wr */ { BxTraceEnd, BX_IA_HLT },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F4 /wm */ { BxTraceEnd, BX_IA_HLT },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F5 /wr */ { 0, BX_IA_CMC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F5 /wm */ { 0, BX_IA_CMC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F6 /wr */ { BxGroup3, BX_IA_ERROR, BxOpcodeInfoG3EbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F6 /wm */ { BxGroup3, BX_IA_ERROR, BxOpcodeInfoG3EbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F7 /wr */ { BxGroup3, BX_IA_ERROR, BxOpcodeInfoG3EwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F7 /wm */ { BxGroup3, BX_IA_ERROR, BxOpcodeInfoG3EwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F8 /wr */ { 0, BX_IA_CLC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F8 /wm */ { 0, BX_IA_CLC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F9 /wr */ { 0, BX_IA_STC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F9 /wm */ { 0, BX_IA_STC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* FA /wr */ { 0, BX_IA_CLI },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* FA /wm */ { 0, BX_IA_CLI },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* FB /wr */ { 0, BX_IA_STI },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* FB /wm */ { 0, BX_IA_STI },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* FC /wr */ { 0, BX_IA_CLD },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* FC /wm */ { 0, BX_IA_CLD },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* FD /wr */ { 0, BX_IA_STD },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* FD /wm */ { 0, BX_IA_STD },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* FE /wr */ { BxGroup4, BX_IA_ERROR, BxOpcodeInfoG4R },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* FE /wm */ { BxGroup4, BX_IA_ERROR, BxOpcodeInfoG4M },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* FF /wr */ { BxGroup5, BX_IA_ERROR, BxOpcodeInfoG5wR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* FF /wm */ { BxGroup5, BX_IA_ERROR, BxOpcodeInfoG5wM },
Fixed CALL/JMP far through call gate 64 Decode SWAPGS and RDTSCP instructions Indent changes in fetchdecode
2006-03-22 23:47:11 +03:00
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 00 /wr */ { BxGroup6, BX_IA_ERROR, BxOpcodeInfoG6 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 00 /wm */ { BxGroup6, BX_IA_ERROR, BxOpcodeInfoG6 },
Fixed XSAVE decode in x86-64 mode
2009-01-10 19:01:55 +03:00
/* 0F 01 /wr */ { BxFPEscape, BX_IA_ERROR, BxOpcodeInfoG7R },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 01 /wm */ { BxGroup7, BX_IA_ERROR, BxOpcodeInfoG7M },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 02 /wr */ { 0, BX_IA_LAR_GvEw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 02 /wm */ { 0, BX_IA_LAR_GvEw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 03 /wr */ { 0, BX_IA_LSL_GvEw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 03 /wm */ { 0, BX_IA_LSL_GvEw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 04 /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 04 /wm */ { 0, BX_IA_ERROR },
Implemented SYSCALL and SYSRET as part of x86-64 emulation. Since the SYSCALL replaces the LOADALL instruction, it is incompatible with earlier CPU types. At moment, the SYSCALL is only enabled by x86-64 emulation, but the code can be incorporated in IA32 only emulations. Instructions added: 0F 05 SYSCALL (replaces LOADALL) 0F 07 SYSRET (new) TODO: restructure #if ... so that it can be used by non x86-64 emulations.
2002-09-25 16:54:41 +04:00
#if BX_SUPPORT_X86_64
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 05 /wr */ { BxTraceEnd, BX_IA_SYSCALL },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 05 /wm */ { BxTraceEnd, BX_IA_SYSCALL },
LOADALL for cpu-level=2 in fetchdecode
2004-05-11 20:44:58 +04:00
#elif BX_CPU_LEVEL == 2
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 05 /wr */ { BxTraceEnd, BX_IA_LOADALL },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 05 /wm */ { BxTraceEnd, BX_IA_LOADALL },
Implemented SYSCALL and SYSRET as part of x86-64 emulation. Since the SYSCALL replaces the LOADALL instruction, it is incompatible with earlier CPU types. At moment, the SYSCALL is only enabled by x86-64 emulation, but the code can be incorporated in IA32 only emulations. Instructions added: 0F 05 SYSCALL (replaces LOADALL) 0F 07 SYSRET (new) TODO: restructure #if ... so that it can be used by non x86-64 emulations.
2002-09-25 16:54:41 +04:00
#else
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 05 /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 05 /wm */ { 0, BX_IA_ERROR },
Implemented SYSCALL and SYSRET as part of x86-64 emulation. Since the SYSCALL replaces the LOADALL instruction, it is incompatible with earlier CPU types. At moment, the SYSCALL is only enabled by x86-64 emulation, but the code can be incorporated in IA32 only emulations. Instructions added: 0F 05 SYSCALL (replaces LOADALL) 0F 07 SYSRET (new) TODO: restructure #if ... so that it can be used by non x86-64 emulations.
2002-09-25 16:54:41 +04:00
#endif
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 06 /wr */ { 0, BX_IA_CLTS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 06 /wm */ { 0, BX_IA_CLTS },
Implemented SYSCALL and SYSRET as part of x86-64 emulation. Since the SYSCALL replaces the LOADALL instruction, it is incompatible with earlier CPU types. At moment, the SYSCALL is only enabled by x86-64 emulation, but the code can be incorporated in IA32 only emulations. Instructions added: 0F 05 SYSCALL (replaces LOADALL) 0F 07 SYSRET (new) TODO: restructure #if ... so that it can be used by non x86-64 emulations.
2002-09-25 16:54:41 +04:00
#if BX_SUPPORT_X86_64
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 07 /wr */ { BxTraceEnd, BX_IA_SYSRET },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 07 /wm */ { BxTraceEnd, BX_IA_SYSRET },
Implemented SYSCALL and SYSRET as part of x86-64 emulation. Since the SYSCALL replaces the LOADALL instruction, it is incompatible with earlier CPU types. At moment, the SYSCALL is only enabled by x86-64 emulation, but the code can be incorporated in IA32 only emulations. Instructions added: 0F 05 SYSCALL (replaces LOADALL) 0F 07 SYSRET (new) TODO: restructure #if ... so that it can be used by non x86-64 emulations.
2002-09-25 16:54:41 +04:00
#else
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 07 /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 07 /wm */ { 0, BX_IA_ERROR },
Implemented SYSCALL and SYSRET as part of x86-64 emulation. Since the SYSCALL replaces the LOADALL instruction, it is incompatible with earlier CPU types. At moment, the SYSCALL is only enabled by x86-64 emulation, but the code can be incorporated in IA32 only emulations. Instructions added: 0F 05 SYSCALL (replaces LOADALL) 0F 07 SYSRET (new) TODO: restructure #if ... so that it can be used by non x86-64 emulations.
2002-09-25 16:54:41 +04:00
#endif
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 08 /wr */ { BxTraceEnd, BX_IA_INVD },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 08 /wm */ { BxTraceEnd, BX_IA_INVD },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 09 /wr */ { BxTraceEnd, BX_IA_WBINVD },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 09 /wm */ { BxTraceEnd, BX_IA_WBINVD },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 0A /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 0A /wm */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 0B /wr */ { BxTraceEnd, BX_IA_UD2A },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 0B /wm */ { BxTraceEnd, BX_IA_UD2A },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 0C /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 0C /wm */ { 0, BX_IA_ERROR },
1st step of moving CPU options to runtime
2010-02-24 22:27:51 +03:00
/* 0F 0D /wr */ { 0, BX_IA_PREFETCHW }, // 3DNow! PREFETCH on AMD, NOP on Intel
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 0D /wm */ { 0, BX_IA_PREFETCHW }, // 3DNow! PREFETCH on AMD, NOP on Intel
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 0E /wr */ { 0, BX_IA_FEMMS }, // 3DNow! FEMMS
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 0E /wm */ { 0, BX_IA_FEMMS }, // 3DNow! FEMMS
Merged NEW-INSTRUCTIONS branch
2003-05-15 20:41:17 +04:00
#if BX_SUPPORT_3DNOW
compilation fix
2010-04-30 01:04:23 +04:00
/* 0F 0F /wr */ { BxImmediate_Ib, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 0F /wm */ { BxImmediate_Ib, BX_IA_ERROR },
Merged NEW-INSTRUCTIONS branch
2003-05-15 20:41:17 +04:00
#else
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 0F /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 0F /wm */ { 0, BX_IA_ERROR },
Merged NEW-INSTRUCTIONS branch
2003-05-15 20:41:17 +04:00
#endif
split SSE move instructions
2009-10-24 15:17:51 +04:00
/* 0F 10 /wr */ { BxPrefixSSE, BX_IA_MOVUPS_VpsWpsR, BxOpcodeGroupSSE_0f10R },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 10 /wm */ { BxPrefixSSE, BX_IA_MOVUPS_VpsWpsM, BxOpcodeGroupSSE_0f10M },
split SSE move instructions
2009-10-24 15:17:51 +04:00
/* 0F 11 /wr */ { BxPrefixSSE, BX_IA_MOVUPS_WpsVpsR, BxOpcodeGroupSSE_0f11R },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 11 /wm */ { BxPrefixSSE, BX_IA_MOVUPS_WpsVpsM, BxOpcodeGroupSSE_0f11M },
split 2 more SSE opcodes
2010-04-08 21:35:32 +04:00
/* 0F 12 /wr */ { BxPrefixSSE, BX_IA_MOVHLPS_VpsWpsR, BxOpcodeGroupSSE_0f12R },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 12 /wm */ { BxPrefixSSE, BX_IA_MOVLPS_VpsMq, BxOpcodeGroupSSE_0f12M },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 13 /wr */ { 0, BX_IA_ERROR }, // MOVLPS/PD SSE group
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 13 /wm */ { BxPrefixSSE, BX_IA_MOVLPS_MqVps, BxOpcodeGroupSSE_0f13M },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 14 /wr */ { BxPrefixSSE, BX_IA_UNPCKLPS_VpsWdq, BxOpcodeGroupSSE_0f14 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 14 /wm */ { BxPrefixSSE, BX_IA_UNPCKLPS_VpsWdq, BxOpcodeGroupSSE_0f14 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 15 /wr */ { BxPrefixSSE, BX_IA_UNPCKHPS_VpsWdq, BxOpcodeGroupSSE_0f15 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 15 /wm */ { BxPrefixSSE, BX_IA_UNPCKHPS_VpsWdq, BxOpcodeGroupSSE_0f15 },
split 2 more SSE opcodes
2010-04-08 21:35:32 +04:00
/* 0F 16 /wr */ { BxPrefixSSE, BX_IA_MOVLHPS_VpsWpsR, BxOpcodeGroupSSE_0f16R },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 16 /wm */ { BxPrefixSSE, BX_IA_MOVHPS_VpsMq, BxOpcodeGroupSSE_0f16M },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 17 /wr */ { 0, BX_IA_ERROR }, // MOHLPS/PD SSE group
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 17 /wm */ { BxPrefixSSE, BX_IA_MOVHPS_MqVps, BxOpcodeGroupSSE_0f17M },
- prefetch hint should be NOP when use in register mode - #GP when trying to set reserved bits of CR4_HI in 64-bit mode - #GP when trying to set reserved bits of EFER MSR - clear upper part of RSI/RDI when executing rep instructions with 32-bit asize even if no repeat iterations were executed (because of RCX=0 for example) - write SYSENTER_EIP_MSR and SYSENTER_ESP_MSR as 64-bit when x86_64 supported - set MSR_FMASK reset value - MSR_FMASK should be 32-bit only - check for fetch permissions when doing ITLB lookup - #GP when trying to write non-canonical address to MSR_CSTAR or MSR_LSTAR - correct repeat instructions timing - mark TSS busy in TR after it is loaded
2008-04-16 20:44:06 +04:00
#if BX_CPU_LEVEL >= 6
/* 0F 18 /wr */ { 0, BX_IA_NOP }, // PREFETCH HINT
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 18 /wm */ { 0, BX_IA_PREFETCH }, // opcode group G16, PREFETCH hints
- prefetch hint should be NOP when use in register mode - #GP when trying to set reserved bits of CR4_HI in 64-bit mode - #GP when trying to set reserved bits of EFER MSR - clear upper part of RSI/RDI when executing rep instructions with 32-bit asize even if no repeat iterations were executed (because of RCX=0 for example) - write SYSENTER_EIP_MSR and SYSENTER_ESP_MSR as 64-bit when x86_64 supported - set MSR_FMASK reset value - MSR_FMASK should be 32-bit only - check for fetch permissions when doing ITLB lookup - #GP when trying to write non-canonical address to MSR_CSTAR or MSR_LSTAR - correct repeat instructions timing - mark TSS busy in TR after it is loaded
2008-04-16 20:44:06 +04:00
/* 0F 19 /wr */ { 0, BX_IA_NOP }, // multi-byte NOP
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 19 /wm */ { 0, BX_IA_NOP }, // multi-byte NOP
- prefetch hint should be NOP when use in register mode - #GP when trying to set reserved bits of CR4_HI in 64-bit mode - #GP when trying to set reserved bits of EFER MSR - clear upper part of RSI/RDI when executing rep instructions with 32-bit asize even if no repeat iterations were executed (because of RCX=0 for example) - write SYSENTER_EIP_MSR and SYSENTER_ESP_MSR as 64-bit when x86_64 supported - set MSR_FMASK reset value - MSR_FMASK should be 32-bit only - check for fetch permissions when doing ITLB lookup - #GP when trying to write non-canonical address to MSR_CSTAR or MSR_LSTAR - correct repeat instructions timing - mark TSS busy in TR after it is loaded
2008-04-16 20:44:06 +04:00
/* 0F 1A /wr */ { 0, BX_IA_NOP }, // multi-byte NOP
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1A /wm */ { 0, BX_IA_NOP }, // multi-byte NOP
- prefetch hint should be NOP when use in register mode - #GP when trying to set reserved bits of CR4_HI in 64-bit mode - #GP when trying to set reserved bits of EFER MSR - clear upper part of RSI/RDI when executing rep instructions with 32-bit asize even if no repeat iterations were executed (because of RCX=0 for example) - write SYSENTER_EIP_MSR and SYSENTER_ESP_MSR as 64-bit when x86_64 supported - set MSR_FMASK reset value - MSR_FMASK should be 32-bit only - check for fetch permissions when doing ITLB lookup - #GP when trying to write non-canonical address to MSR_CSTAR or MSR_LSTAR - correct repeat instructions timing - mark TSS busy in TR after it is loaded
2008-04-16 20:44:06 +04:00
/* 0F 1B /wr */ { 0, BX_IA_NOP }, // multi-byte NOP
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1B /wm */ { 0, BX_IA_NOP }, // multi-byte NOP
- prefetch hint should be NOP when use in register mode - #GP when trying to set reserved bits of CR4_HI in 64-bit mode - #GP when trying to set reserved bits of EFER MSR - clear upper part of RSI/RDI when executing rep instructions with 32-bit asize even if no repeat iterations were executed (because of RCX=0 for example) - write SYSENTER_EIP_MSR and SYSENTER_ESP_MSR as 64-bit when x86_64 supported - set MSR_FMASK reset value - MSR_FMASK should be 32-bit only - check for fetch permissions when doing ITLB lookup - #GP when trying to write non-canonical address to MSR_CSTAR or MSR_LSTAR - correct repeat instructions timing - mark TSS busy in TR after it is loaded
2008-04-16 20:44:06 +04:00
/* 0F 1C /wr */ { 0, BX_IA_NOP }, // multi-byte NOP
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1C /wm */ { 0, BX_IA_NOP }, // multi-byte NOP
- prefetch hint should be NOP when use in register mode - #GP when trying to set reserved bits of CR4_HI in 64-bit mode - #GP when trying to set reserved bits of EFER MSR - clear upper part of RSI/RDI when executing rep instructions with 32-bit asize even if no repeat iterations were executed (because of RCX=0 for example) - write SYSENTER_EIP_MSR and SYSENTER_ESP_MSR as 64-bit when x86_64 supported - set MSR_FMASK reset value - MSR_FMASK should be 32-bit only - check for fetch permissions when doing ITLB lookup - #GP when trying to write non-canonical address to MSR_CSTAR or MSR_LSTAR - correct repeat instructions timing - mark TSS busy in TR after it is loaded
2008-04-16 20:44:06 +04:00
/* 0F 1D /wr */ { 0, BX_IA_NOP }, // multi-byte NOP
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1D /wm */ { 0, BX_IA_NOP }, // multi-byte NOP
- prefetch hint should be NOP when use in register mode - #GP when trying to set reserved bits of CR4_HI in 64-bit mode - #GP when trying to set reserved bits of EFER MSR - clear upper part of RSI/RDI when executing rep instructions with 32-bit asize even if no repeat iterations were executed (because of RCX=0 for example) - write SYSENTER_EIP_MSR and SYSENTER_ESP_MSR as 64-bit when x86_64 supported - set MSR_FMASK reset value - MSR_FMASK should be 32-bit only - check for fetch permissions when doing ITLB lookup - #GP when trying to write non-canonical address to MSR_CSTAR or MSR_LSTAR - correct repeat instructions timing - mark TSS busy in TR after it is loaded
2008-04-16 20:44:06 +04:00
/* 0F 1E /wr */ { 0, BX_IA_NOP }, // multi-byte NOP
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1E /wm */ { 0, BX_IA_NOP }, // multi-byte NOP
- prefetch hint should be NOP when use in register mode - #GP when trying to set reserved bits of CR4_HI in 64-bit mode - #GP when trying to set reserved bits of EFER MSR - clear upper part of RSI/RDI when executing rep instructions with 32-bit asize even if no repeat iterations were executed (because of RCX=0 for example) - write SYSENTER_EIP_MSR and SYSENTER_ESP_MSR as 64-bit when x86_64 supported - set MSR_FMASK reset value - MSR_FMASK should be 32-bit only - check for fetch permissions when doing ITLB lookup - #GP when trying to write non-canonical address to MSR_CSTAR or MSR_LSTAR - correct repeat instructions timing - mark TSS busy in TR after it is loaded
2008-04-16 20:44:06 +04:00
/* 0F 1F /wr */ { 0, BX_IA_NOP }, // multi-byte NOP
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1F /wm */ { 0, BX_IA_NOP }, // multi-byte NOP
- prefetch hint should be NOP when use in register mode - #GP when trying to set reserved bits of CR4_HI in 64-bit mode - #GP when trying to set reserved bits of EFER MSR - clear upper part of RSI/RDI when executing rep instructions with 32-bit asize even if no repeat iterations were executed (because of RCX=0 for example) - write SYSENTER_EIP_MSR and SYSENTER_ESP_MSR as 64-bit when x86_64 supported - set MSR_FMASK reset value - MSR_FMASK should be 32-bit only - check for fetch permissions when doing ITLB lookup - #GP when trying to write non-canonical address to MSR_CSTAR or MSR_LSTAR - correct repeat instructions timing - mark TSS busy in TR after it is loaded
2008-04-16 20:44:06 +04:00
#else
/* 0F 18 /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 18 /wm */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 19 /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 19 /wm */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 1A /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1A /wm */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 1B /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1B /wm */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 1C /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1C /wm */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 1D /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1D /wm */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 1E /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1E /wm */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 1F /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1F /wm */ { 0, BX_IA_ERROR },
Changes list made after CVS service crash: - Fixed critical bug in CPU code added with one of the prev commits - Disasm support for SSE4 - Rename PNI->SSE3 everywhere in the code - Correctly decode, disassemble and execute 'XCHG R8, rAX' x86-64 instruction - Correctly decode, disassemble and execute multi-byte NOP 0F F1 opcode - Fixed ENTER and LEAVE instructions in x86-64 mode - Added ability to turn ON instruction trace, only GUI support is missed. Instruction trace could be enabled if Bochs was compiled with disasm - More changes Bit32u -> bx_phy_address - Complete preliminary implementation of SMM in Bochs, SMI is still PANICs but if you press 'continue' everything should work OK - Small code cleanup - Update CHANGES and user docs
2006-04-05 21:31:35 +04:00
#endif
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 20 /wr */ { 0, BX_IA_MOV_RdCd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 20 /wm */ { 0, BX_IA_MOV_RdCd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 21 /wr */ { 0, BX_IA_MOV_RdDd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 21 /wm */ { 0, BX_IA_MOV_RdDd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 22 /wr */ { BxTraceEnd, BX_IA_MOV_CdRd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 22 /wm */ { BxTraceEnd, BX_IA_MOV_CdRd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 23 /wr */ { BxTraceEnd, BX_IA_MOV_DdRd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 23 /wm */ { BxTraceEnd, BX_IA_MOV_DdRd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 24 /wr */ { BxTraceEnd, BX_IA_MOV_RdTd }, // not implemented
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 24 /wm */ { BxTraceEnd, BX_IA_MOV_RdTd }, // not implemented
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 25 /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 25 /wm */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 26 /wr */ { BxTraceEnd, BX_IA_MOV_TdRd }, // not implemented
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 26 /wm */ { BxTraceEnd, BX_IA_MOV_TdRd }, // not implemented
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 27 /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 27 /wm */ { 0, BX_IA_ERROR },
split SSE move instructions
2009-10-24 15:17:51 +04:00
/* 0F 28 /wr */ { BxPrefixSSE, BX_IA_MOVAPS_VpsWpsR, BxOpcodeGroupSSE_0f28R },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 28 /wm */ { BxPrefixSSE, BX_IA_MOVAPS_VpsWpsM, BxOpcodeGroupSSE_0f28M },
split SSE move instructions
2009-10-24 15:17:51 +04:00
/* 0F 29 /wr */ { BxPrefixSSE, BX_IA_MOVAPS_WpsVpsR, BxOpcodeGroupSSE_0f29R },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 29 /wm */ { BxPrefixSSE, BX_IA_MOVAPS_WpsVpsM, BxOpcodeGroupSSE_0f29M },
split opcodes by ModC0
2009-08-22 15:47:42 +04:00
/* 0F 2A /wr */ { BxPrefixSSE, BX_IA_CVTPI2PS_VpsQqR, BxOpcodeGroupSSE_0f2aR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 2A /wm */ { BxPrefixSSE, BX_IA_CVTPI2PS_VpsQqM, BxOpcodeGroupSSE_0f2aM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 2B /wr */ { 0, BX_IA_ERROR }, // MOVNTPS/PD/SS/SD
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 2B /wm */ { BxPrefixSSE, BX_IA_MOVNTPS_MpsVps, BxOpcodeGroupSSE_0f2bM },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 2C /wr */ { BxPrefixSSE, BX_IA_CVTTPS2PI_PqWps, BxOpcodeGroupSSE_0f2c },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 2C /wm */ { BxPrefixSSE, BX_IA_CVTTPS2PI_PqWps, BxOpcodeGroupSSE_0f2c },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 2D /wr */ { BxPrefixSSE, BX_IA_CVTPS2PI_PqWps, BxOpcodeGroupSSE_0f2d },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 2D /wm */ { BxPrefixSSE, BX_IA_CVTPS2PI_PqWps, BxOpcodeGroupSSE_0f2d },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 2E /wr */ { BxPrefixSSE, BX_IA_UCOMISS_VssWss, BxOpcodeGroupSSE_0f2e },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 2E /wm */ { BxPrefixSSE, BX_IA_UCOMISS_VssWss, BxOpcodeGroupSSE_0f2e },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 2F /wr */ { BxPrefixSSE, BX_IA_COMISS_VpsWps, BxOpcodeGroupSSE_0f2f },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 2F /wm */ { BxPrefixSSE, BX_IA_COMISS_VpsWps, BxOpcodeGroupSSE_0f2f },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 30 /wr */ { 0, BX_IA_WRMSR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 30 /wm */ { 0, BX_IA_WRMSR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 31 /wr */ { 0, BX_IA_RDTSC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 31 /wm */ { 0, BX_IA_RDTSC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 32 /wr */ { 0, BX_IA_RDMSR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 32 /wm */ { 0, BX_IA_RDMSR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 33 /wr */ { 0, BX_IA_RDPMC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 33 /wm */ { 0, BX_IA_RDPMC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 34 /wr */ { BxTraceEnd, BX_IA_SYSENTER },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 34 /wm */ { BxTraceEnd, BX_IA_SYSENTER },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 35 /wr */ { BxTraceEnd, BX_IA_SYSEXIT },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 35 /wm */ { BxTraceEnd, BX_IA_SYSEXIT },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 36 /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 36 /wm */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 37 /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 37 /wm */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 38 /wr */ { Bx3ByteOp, BX_IA_ERROR, BxOpcode3ByteTable0f38 }, // 3-byte escape
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 38 /wm */ { Bx3ByteOp, BX_IA_ERROR, BxOpcode3ByteTable0f38 }, // 3-byte escape
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 39 /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 39 /wm */ { 0, BX_IA_ERROR },
minimize SSE tables, minor speedup in SSE code
2009-02-27 00:57:01 +03:00
/* 0F 3A /wr */ { Bx3ByteOp | BxImmediate_Ib, BX_IA_ERROR, BxOpcode3ByteTable0f3a }, // 3-byte escape
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 3A /wm */ { Bx3ByteOp | BxImmediate_Ib, BX_IA_ERROR, BxOpcode3ByteTable0f3a }, // 3-byte escape
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 3B /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 3B /wm */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 3C /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 3C /wm */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 3D /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 3D /wm */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 3E /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 3E /wm */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 3F /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 3F /wm */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 40 /wr */ { 0, BX_IA_CMOVO_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 40 /wm */ { 0, BX_IA_CMOVO_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 41 /wr */ { 0, BX_IA_CMOVNO_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 41 /wm */ { 0, BX_IA_CMOVNO_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 42 /wr */ { 0, BX_IA_CMOVB_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 42 /wm */ { 0, BX_IA_CMOVB_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 43 /wr */ { 0, BX_IA_CMOVNB_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 43 /wm */ { 0, BX_IA_CMOVNB_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 44 /wr */ { 0, BX_IA_CMOVZ_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 44 /wm */ { 0, BX_IA_CMOVZ_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 45 /wr */ { 0, BX_IA_CMOVNZ_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 45 /wm */ { 0, BX_IA_CMOVNZ_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 46 /wr */ { 0, BX_IA_CMOVBE_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 46 /wm */ { 0, BX_IA_CMOVBE_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 47 /wr */ { 0, BX_IA_CMOVNBE_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 47 /wm */ { 0, BX_IA_CMOVNBE_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 48 /wr */ { 0, BX_IA_CMOVS_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 48 /wm */ { 0, BX_IA_CMOVS_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 49 /wr */ { 0, BX_IA_CMOVNS_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 49 /wm */ { 0, BX_IA_CMOVNS_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 4A /wr */ { 0, BX_IA_CMOVP_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 4A /wm */ { 0, BX_IA_CMOVP_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 4B /wr */ { 0, BX_IA_CMOVNP_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 4B /wm */ { 0, BX_IA_CMOVNP_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 4C /wr */ { 0, BX_IA_CMOVL_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 4C /wm */ { 0, BX_IA_CMOVL_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 4D /wr */ { 0, BX_IA_CMOVNL_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 4D /wm */ { 0, BX_IA_CMOVNL_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 4E /wr */ { 0, BX_IA_CMOVLE_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 4E /wm */ { 0, BX_IA_CMOVLE_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 4F /wr */ { 0, BX_IA_CMOVNLE_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 4F /wm */ { 0, BX_IA_CMOVNLE_GwEwM },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 50 /wr */ { BxPrefixSSE, BX_IA_MOVMSKPS_GdVRps, BxOpcodeGroupSSE_0f50 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 50 /wm */ { 0, BX_IA_ERROR }, // MOVMSKPS/PD
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 51 /wr */ { BxPrefixSSE, BX_IA_SQRTPS_VpsWps, BxOpcodeGroupSSE_0f51 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 51 /wm */ { BxPrefixSSE, BX_IA_SQRTPS_VpsWps, BxOpcodeGroupSSE_0f51 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 52 /wr */ { BxPrefixSSE, BX_IA_RSQRTPS_VpsWps, BxOpcodeGroupSSE_0f52 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 52 /wm */ { BxPrefixSSE, BX_IA_RSQRTPS_VpsWps, BxOpcodeGroupSSE_0f52 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 53 /wr */ { BxPrefixSSE, BX_IA_RCPPS_VpsWps, BxOpcodeGroupSSE_0f53 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 53 /wm */ { BxPrefixSSE, BX_IA_RCPPS_VpsWps, BxOpcodeGroupSSE_0f53 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 54 /wr */ { BxPrefixSSE, BX_IA_ANDPS_VpsWps, BxOpcodeGroupSSE_0f54 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 54 /wm */ { BxPrefixSSE, BX_IA_ANDPS_VpsWps, BxOpcodeGroupSSE_0f54 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 55 /wr */ { BxPrefixSSE, BX_IA_ANDNPS_VpsWps, BxOpcodeGroupSSE_0f55 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 55 /wm */ { BxPrefixSSE, BX_IA_ANDNPS_VpsWps, BxOpcodeGroupSSE_0f55 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 56 /wr */ { BxPrefixSSE, BX_IA_ORPS_VpsWps, BxOpcodeGroupSSE_0f56 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 56 /wm */ { BxPrefixSSE, BX_IA_ORPS_VpsWps, BxOpcodeGroupSSE_0f56 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 57 /wr */ { BxPrefixSSE, BX_IA_XORPS_VpsWps, BxOpcodeGroupSSE_0f57 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 57 /wm */ { BxPrefixSSE, BX_IA_XORPS_VpsWps, BxOpcodeGroupSSE_0f57 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 58 /wr */ { BxPrefixSSE, BX_IA_ADDPS_VpsWps, BxOpcodeGroupSSE_0f58 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 58 /wm */ { BxPrefixSSE, BX_IA_ADDPS_VpsWps, BxOpcodeGroupSSE_0f58 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 59 /wr */ { BxPrefixSSE, BX_IA_MULPS_VpsWps, BxOpcodeGroupSSE_0f59 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 59 /wm */ { BxPrefixSSE, BX_IA_MULPS_VpsWps, BxOpcodeGroupSSE_0f59 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 5A /wr */ { BxPrefixSSE, BX_IA_CVTPS2PD_VpsWps, BxOpcodeGroupSSE_0f5a },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 5A /wm */ { BxPrefixSSE, BX_IA_CVTPS2PD_VpsWps, BxOpcodeGroupSSE_0f5a },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 5B /wr */ { BxPrefixSSE, BX_IA_CVTDQ2PS_VpsWdq, BxOpcodeGroupSSE_0f5b },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 5B /wm */ { BxPrefixSSE, BX_IA_CVTDQ2PS_VpsWdq, BxOpcodeGroupSSE_0f5b },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 5C /wr */ { BxPrefixSSE, BX_IA_SUBPS_VpsWps, BxOpcodeGroupSSE_0f5c },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 5C /wm */ { BxPrefixSSE, BX_IA_SUBPS_VpsWps, BxOpcodeGroupSSE_0f5c },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 5D /wr */ { BxPrefixSSE, BX_IA_MINPS_VpsWps, BxOpcodeGroupSSE_0f5d },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 5D /wm */ { BxPrefixSSE, BX_IA_MINPS_VpsWps, BxOpcodeGroupSSE_0f5d },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 5E /wr */ { BxPrefixSSE, BX_IA_DIVPS_VpsWps, BxOpcodeGroupSSE_0f5e },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 5E /wm */ { BxPrefixSSE, BX_IA_DIVPS_VpsWps, BxOpcodeGroupSSE_0f5e },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 5F /wr */ { BxPrefixSSE, BX_IA_MAXPS_VpsWps, BxOpcodeGroupSSE_0f5f },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 5F /wm */ { BxPrefixSSE, BX_IA_MAXPS_VpsWps, BxOpcodeGroupSSE_0f5f },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 60 /wr */ { BxPrefixSSE, BX_IA_PUNPCKLBW_PqQd, BxOpcodeGroupSSE_0f60 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 60 /wm */ { BxPrefixSSE, BX_IA_PUNPCKLBW_PqQd, BxOpcodeGroupSSE_0f60 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 61 /wr */ { BxPrefixSSE, BX_IA_PUNPCKLWD_PqQd, BxOpcodeGroupSSE_0f61 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 61 /wm */ { BxPrefixSSE, BX_IA_PUNPCKLWD_PqQd, BxOpcodeGroupSSE_0f61 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 62 /wr */ { BxPrefixSSE, BX_IA_PUNPCKLDQ_PqQd, BxOpcodeGroupSSE_0f62 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 62 /wm */ { BxPrefixSSE, BX_IA_PUNPCKLDQ_PqQd, BxOpcodeGroupSSE_0f62 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 63 /wr */ { BxPrefixSSE, BX_IA_PACKSSWB_PqQq, BxOpcodeGroupSSE_0f63 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 63 /wm */ { BxPrefixSSE, BX_IA_PACKSSWB_PqQq, BxOpcodeGroupSSE_0f63 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 64 /wr */ { BxPrefixSSE, BX_IA_PCMPGTB_PqQq, BxOpcodeGroupSSE_0f64 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 64 /wm */ { BxPrefixSSE, BX_IA_PCMPGTB_PqQq, BxOpcodeGroupSSE_0f64 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 65 /wr */ { BxPrefixSSE, BX_IA_PCMPGTW_PqQq, BxOpcodeGroupSSE_0f65 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 65 /wm */ { BxPrefixSSE, BX_IA_PCMPGTW_PqQq, BxOpcodeGroupSSE_0f65 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 66 /wr */ { BxPrefixSSE, BX_IA_PCMPGTD_PqQq, BxOpcodeGroupSSE_0f66 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 66 /wm */ { BxPrefixSSE, BX_IA_PCMPGTD_PqQq, BxOpcodeGroupSSE_0f66 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 67 /wr */ { BxPrefixSSE, BX_IA_PACKUSWB_PqQq, BxOpcodeGroupSSE_0f67 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 67 /wm */ { BxPrefixSSE, BX_IA_PACKUSWB_PqQq, BxOpcodeGroupSSE_0f67 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 68 /wr */ { BxPrefixSSE, BX_IA_PUNPCKHBW_PqQq, BxOpcodeGroupSSE_0f68 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 68 /wm */ { BxPrefixSSE, BX_IA_PUNPCKHBW_PqQq, BxOpcodeGroupSSE_0f68 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 69 /wr */ { BxPrefixSSE, BX_IA_PUNPCKHWD_PqQq, BxOpcodeGroupSSE_0f69 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 69 /wm */ { BxPrefixSSE, BX_IA_PUNPCKHWD_PqQq, BxOpcodeGroupSSE_0f69 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 6A /wr */ { BxPrefixSSE, BX_IA_PUNPCKHDQ_PqQq, BxOpcodeGroupSSE_0f6a },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 6A /wm */ { BxPrefixSSE, BX_IA_PUNPCKHDQ_PqQq, BxOpcodeGroupSSE_0f6a },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 6B /wr */ { BxPrefixSSE, BX_IA_PACKSSDW_PqQq, BxOpcodeGroupSSE_0f6b },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 6B /wm */ { BxPrefixSSE, BX_IA_PACKSSDW_PqQq, BxOpcodeGroupSSE_0f6b },
cleanup
2010-02-08 17:54:26 +03:00
/* 0F 6C /wr */ { BxPrefixSSE66, BX_IA_PUNPCKLQDQ_VdqWdq },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 6C /wm */ { BxPrefixSSE66, BX_IA_PUNPCKLQDQ_VdqWdq },
cleanup
2010-02-08 17:54:26 +03:00
/* 0F 6D /wr */ { BxPrefixSSE66, BX_IA_PUNPCKHQDQ_VdqWdq },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 6D /wm */ { BxPrefixSSE66, BX_IA_PUNPCKHQDQ_VdqWdq },
split few more opcodes
2010-02-10 20:21:15 +03:00
/* 0F 6E /wr */ { BxPrefixSSE, BX_IA_MOVD_PqEdR, BxOpcodeGroupSSE_0f6eR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 6E /wm */ { BxPrefixSSE, BX_IA_MOVD_PqEdM, BxOpcodeGroupSSE_0f6eM },
Split MOVQ_PqQq opcode
2009-12-14 14:55:42 +03:00
/* 0F 6F /wr */ { BxPrefixSSE, BX_IA_MOVQ_PqQqR, BxOpcodeGroupSSE_0f6fR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 6F /wm */ { BxPrefixSSE, BX_IA_MOVQ_PqQqM, BxOpcodeGroupSSE_0f6fM },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 70 /wr */ { BxPrefixSSE | BxImmediate_Ib, BX_IA_PSHUFW_PqQqIb, BxOpcodeGroupSSE_0f70 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 70 /wm */ { BxPrefixSSE | BxImmediate_Ib, BX_IA_PSHUFW_PqQqIb, BxOpcodeGroupSSE_0f70 },
More fetchdecode fixes
2008-05-01 01:32:33 +04:00
/* 0F 71 /wr */ { BxGroup12, BX_IA_ERROR, BxOpcodeInfoG12R },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 71 /wm */ { 0, BX_IA_ERROR }, // SSE Group G12
More fetchdecode fixes
2008-05-01 01:32:33 +04:00
/* 0F 72 /wr */ { BxGroup13, BX_IA_ERROR, BxOpcodeInfoG13R },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 72 /wm */ { 0, BX_IA_ERROR }, // SSE Group G13
More fetchdecode fixes
2008-05-01 01:32:33 +04:00
/* 0F 73 /wr */ { BxGroup14, BX_IA_ERROR, BxOpcodeInfoG14R },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 73 /wm */ { 0, BX_IA_ERROR }, // SSE Group G14
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 74 /wr */ { BxPrefixSSE, BX_IA_PCMPEQB_PqQq, BxOpcodeGroupSSE_0f74 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 74 /wm */ { BxPrefixSSE, BX_IA_PCMPEQB_PqQq, BxOpcodeGroupSSE_0f74 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 75 /wr */ { BxPrefixSSE, BX_IA_PCMPEQW_PqQq, BxOpcodeGroupSSE_0f75 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 75 /wm */ { BxPrefixSSE, BX_IA_PCMPEQW_PqQq, BxOpcodeGroupSSE_0f75 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 76 /wr */ { BxPrefixSSE, BX_IA_PCMPEQD_PqQq, BxOpcodeGroupSSE_0f76 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 76 /wm */ { BxPrefixSSE, BX_IA_PCMPEQD_PqQq, BxOpcodeGroupSSE_0f76 },
Fixed handling of prefixes for EMMS Small FPU optimization
2008-09-13 00:59:31 +04:00
/* 0F 77 /wr */ { BxPrefixSSE, BX_IA_EMMS, BxOpcodeGroupSSE_ERR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 77 /wm */ { BxPrefixSSE, BX_IA_EMMS, BxOpcodeGroupSSE_ERR },
Added Intel VMX emulation to Bochs CPU
2009-01-31 13:43:24 +03:00
/* 0F 78 /wr */ { BxPrefixSSE, BX_IA_VMREAD_EdGd, BxOpcodeGroupSSE_ERR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 78 /wm */ { BxPrefixSSE, BX_IA_VMREAD_EdGd, BxOpcodeGroupSSE_ERR },
Added Intel VMX emulation to Bochs CPU
2009-01-31 13:43:24 +03:00
/* 0F 79 /wr */ { BxPrefixSSE, BX_IA_VMWRITE_GdEd, BxOpcodeGroupSSE_ERR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 79 /wm */ { BxPrefixSSE, BX_IA_VMWRITE_GdEd, BxOpcodeGroupSSE_ERR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 7A /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 7A /wm */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 7B /wr */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 7B /wm */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 7C /wr */ { BxPrefixSSE, BX_IA_ERROR, BxOpcodeGroupSSE_0f7c },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 7C /wm */ { BxPrefixSSE, BX_IA_ERROR, BxOpcodeGroupSSE_0f7c },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 7D /wr */ { BxPrefixSSE, BX_IA_ERROR, BxOpcodeGroupSSE_0f7d },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 7D /wm */ { BxPrefixSSE, BX_IA_ERROR, BxOpcodeGroupSSE_0f7d },
split few more opcodes
2010-02-10 20:21:15 +03:00
/* 0F 7E /wr */ { BxPrefixSSE, BX_IA_MOVD_EdPdR, BxOpcodeGroupSSE_0f7eR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 7E /wm */ { BxPrefixSSE, BX_IA_MOVD_EdPdM, BxOpcodeGroupSSE_0f7eM },
split few more opcodes
2010-02-10 20:21:15 +03:00
/* 0F 7F /wr */ { BxPrefixSSE, BX_IA_MOVQ_QqPqR, BxOpcodeGroupSSE_0f7fR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 7F /wm */ { BxPrefixSSE, BX_IA_MOVQ_QqPqM, BxOpcodeGroupSSE_0f7fM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 80 /wr */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JO_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 80 /wm */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JO_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 81 /wr */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JNO_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 81 /wm */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JNO_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 82 /wr */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JB_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 82 /wm */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JB_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 83 /wr */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JNB_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 83 /wm */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JNB_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 84 /wr */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JZ_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 84 /wm */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JZ_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 85 /wr */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JNZ_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 85 /wm */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JNZ_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 86 /wr */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JBE_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 86 /wm */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JBE_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 87 /wr */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JNBE_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 87 /wm */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JNBE_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 88 /wr */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JS_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 88 /wm */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JS_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 89 /wr */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JNS_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 89 /wm */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JNS_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 8A /wr */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JP_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 8A /wm */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JP_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 8B /wr */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JNP_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 8B /wm */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JNP_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 8C /wr */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JL_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 8C /wm */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JL_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 8D /wr */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JNL_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 8D /wm */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JNL_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 8E /wr */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JLE_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 8E /wm */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JLE_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 8F /wr */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JNLE_Jw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 8F /wm */ { BxImmediate_BrOff16 | BxTraceJCC, BX_IA_JNLE_Jw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 90 /wr */ { 0, BX_IA_SETO_EbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 90 /wm */ { 0, BX_IA_SETO_EbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 91 /wr */ { 0, BX_IA_SETNO_EbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 91 /wm */ { 0, BX_IA_SETNO_EbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 92 /wr */ { 0, BX_IA_SETB_EbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 92 /wm */ { 0, BX_IA_SETB_EbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 93 /wr */ { 0, BX_IA_SETNB_EbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 93 /wm */ { 0, BX_IA_SETNB_EbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 94 /wr */ { 0, BX_IA_SETZ_EbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 94 /wm */ { 0, BX_IA_SETZ_EbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 95 /wr */ { 0, BX_IA_SETNZ_EbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 95 /wm */ { 0, BX_IA_SETNZ_EbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 96 /wr */ { 0, BX_IA_SETBE_EbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 96 /wm */ { 0, BX_IA_SETBE_EbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 97 /wr */ { 0, BX_IA_SETNBE_EbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 97 /wm */ { 0, BX_IA_SETNBE_EbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 98 /wr */ { 0, BX_IA_SETS_EbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 98 /wm */ { 0, BX_IA_SETS_EbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 99 /wr */ { 0, BX_IA_SETNS_EbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 99 /wm */ { 0, BX_IA_SETNS_EbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 9A /wr */ { 0, BX_IA_SETP_EbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 9A /wm */ { 0, BX_IA_SETP_EbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 9B /wr */ { 0, BX_IA_SETNP_EbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 9B /wm */ { 0, BX_IA_SETNP_EbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 9C /wr */ { 0, BX_IA_SETL_EbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 9C /wm */ { 0, BX_IA_SETL_EbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 9D /wr */ { 0, BX_IA_SETNL_EbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 9D /wm */ { 0, BX_IA_SETNL_EbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 9E /wr */ { 0, BX_IA_SETLE_EbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 9E /wm */ { 0, BX_IA_SETLE_EbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 9F /wr */ { 0, BX_IA_SETNLE_EbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 9F /wm */ { 0, BX_IA_SETNLE_EbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F A0 /wr */ { 0, BX_IA_PUSH16_FS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A0 /wm */ { 0, BX_IA_PUSH16_FS },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F A1 /wr */ { 0, BX_IA_POP16_FS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A1 /wm */ { 0, BX_IA_POP16_FS },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F A2 /wr */ { 0, BX_IA_CPUID },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A2 /wm */ { 0, BX_IA_CPUID },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F A3 /wr */ { 0, BX_IA_BT_EwGwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A3 /wm */ { 0, BX_IA_BT_EwGwM },
Split SHRD/SHLD instructions
2008-04-05 23:08:01 +04:00
/* 0F A4 /wr */ { BxImmediate_Ib, BX_IA_SHLD_EwGwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A4 /wm */ { BxImmediate_Ib, BX_IA_SHLD_EwGwM },
Split SHRD/SHLD instructions
2008-04-05 23:08:01 +04:00
/* 0F A5 /wr */ { 0, BX_IA_SHLD_EwGwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A5 /wm */ { 0, BX_IA_SHLD_EwGwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F A6 /wr */ { BxTraceEnd, BX_IA_CMPXCHG_XBTS }, // not implemented
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A6 /wm */ { BxTraceEnd, BX_IA_CMPXCHG_XBTS }, // not implemented
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F A7 /wr */ { BxTraceEnd, BX_IA_CMPXCHG_IBTS }, // not implemented
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A7 /wm */ { BxTraceEnd, BX_IA_CMPXCHG_IBTS }, // not implemented
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F A8 /wr */ { 0, BX_IA_PUSH16_GS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A8 /wm */ { 0, BX_IA_PUSH16_GS },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F A9 /wr */ { 0, BX_IA_POP16_GS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A9 /wm */ { 0, BX_IA_POP16_GS },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F AA /wr */ { BxTraceEnd, BX_IA_RSM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F AA /wm */ { BxTraceEnd, BX_IA_RSM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F AB /wr */ { 0, BX_IA_BTS_EwGwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F AB /wm */ { BxLockable, BX_IA_BTS_EwGwM },
Split SHRD/SHLD instructions
2008-04-05 23:08:01 +04:00
/* 0F AC /wr */ { BxImmediate_Ib, BX_IA_SHRD_EwGwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F AC /wm */ { BxImmediate_Ib, BX_IA_SHRD_EwGwM },
Split SHRD/SHLD instructions
2008-04-05 23:08:01 +04:00
/* 0F AD /wr */ { 0, BX_IA_SHRD_EwGwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F AD /wm */ { 0, BX_IA_SHRD_EwGwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F AE /wr */ { BxGroup15, BX_IA_ERROR, BxOpcodeInfoG15R },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F AE /wm */ { BxGroup15, BX_IA_ERROR, BxOpcodeInfoG15M },
Split more opcodes using new LOAD technique
2008-08-11 01:16:12 +04:00
/* 0F AF /wr */ { 0, BX_IA_IMUL_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F AF /wm */ { 0, BX_IA_IMUL_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F B0 /wr */ { 0, BX_IA_CMPXCHG_EbGbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B0 /wm */ { BxLockable, BX_IA_CMPXCHG_EbGbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F B1 /wr */ { 0, BX_IA_CMPXCHG_EwGwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B1 /wm */ { BxLockable, BX_IA_CMPXCHG_EwGwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F B2 /wr */ { 0, BX_IA_ERROR }, // LSS
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B2 /wm */ { 0, BX_IA_LSS_GwMp },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F B3 /wr */ { 0, BX_IA_BTR_EwGwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B3 /wm */ { BxLockable, BX_IA_BTR_EwGwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F B4 /wr */ { 0, BX_IA_ERROR }, // LFS
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B4 /wm */ { 0, BX_IA_LFS_GwMp },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F B5 /wr */ { 0, BX_IA_ERROR }, // LGS
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B5 /wm */ { 0, BX_IA_LGS_GwMp },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F B6 /wr */ { 0, BX_IA_MOVZX_GwEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B6 /wm */ { 0, BX_IA_MOVZX_GwEbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F B7 /wr */ { 0, BX_IA_MOV_GwEwR }, // MOVZX_GwEw
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B7 /wm */ { 0, BX_IA_MOV_GwEwM }, // MOVZX_GwEw
cleanup
2010-02-08 17:54:26 +03:00
/* 0F B8 /wr */ { BxPrefixSSEF3, BX_IA_POPCNT_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B8 /wm */ { BxPrefixSSEF3, BX_IA_POPCNT_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F B9 /wr */ { 0, BX_IA_UD2B },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B9 /wm */ { BxTraceEnd, BX_IA_UD2B },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F BA /wr */ { BxGroup8, BX_IA_ERROR, BxOpcodeInfoG8EwIbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F BA /wm */ { BxGroup8, BX_IA_ERROR, BxOpcodeInfoG8EwIbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F BB /wr */ { 0, BX_IA_BTC_EwGwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F BB /wm */ { BxLockable, BX_IA_BTC_EwGwM },
More use of LOAD_Ex method
2008-08-10 23:34:28 +04:00
/* 0F BC /wr */ { 0, BX_IA_BSF_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F BC /wm */ { 0, BX_IA_BSF_GwEwM },
More use of LOAD_Ex method
2008-08-10 23:34:28 +04:00
/* 0F BD /wr */ { 0, BX_IA_BSR_GwEwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F BD /wm */ { 0, BX_IA_BSR_GwEwM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F BE /wr */ { 0, BX_IA_MOVSX_GwEbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F BE /wm */ { 0, BX_IA_MOVSX_GwEbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F BF /wr */ { 0, BX_IA_MOV_GwEwR }, // MOVSX_GwEw
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F BF /wm */ { 0, BX_IA_MOV_GwEwM }, // MOVSX_GwEw
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F C0 /wr */ { 0, BX_IA_XADD_EbGbR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C0 /wm */ { BxLockable, BX_IA_XADD_EbGbM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F C1 /wr */ { 0, BX_IA_XADD_EwGwR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C1 /wm */ { BxLockable, BX_IA_XADD_EwGwM },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F C2 /wr */ { BxPrefixSSE | BxImmediate_Ib, BX_IA_CMPPS_VpsWpsIb, BxOpcodeGroupSSE_0fc2 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C2 /wm */ { BxPrefixSSE | BxImmediate_Ib, BX_IA_CMPPS_VpsWpsIb, BxOpcodeGroupSSE_0fc2 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F C3 /wr */ { 0, BX_IA_ERROR }, // MOVNTI SSE group
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C3 /wm */ { BxPrefixSSE, BX_IA_MOVNTI_MdGd, BxOpcodeGroupSSE_ERR },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F C4 /wr */ { BxPrefixSSE | BxImmediate_Ib, BX_IA_PINSRW_PqEwIb, BxOpcodeGroupSSE_0fc4 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C4 /wm */ { BxPrefixSSE | BxImmediate_Ib, BX_IA_PINSRW_PqEwIb, BxOpcodeGroupSSE_0fc4 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F C5 /wr */ { BxPrefixSSE | BxImmediate_Ib, BX_IA_PEXTRW_GdPqIb, BxOpcodeGroupSSE_0fc5R },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C5 /wm */ { 0, BX_IA_ERROR }, // PEXTRW SSE group
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F C6 /wr */ { BxPrefixSSE | BxImmediate_Ib, BX_IA_SHUFPS_VpsWpsIb, BxOpcodeGroupSSE_0fc6 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C6 /wm */ { BxPrefixSSE | BxImmediate_Ib, BX_IA_SHUFPS_VpsWpsIb, BxOpcodeGroupSSE_0fc6 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F C7 /wr */ { 0, BX_IA_ERROR }, // CMPXCHG8B G9 group
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C7 /wm */ { BxGroup9, BX_IA_ERROR, BxOpcodeInfoG9M },
bswap undefined behavior
2010-03-19 13:00:48 +03:00
/* 0F C8 /wr */ { 0, BX_IA_BSWAP_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C8 /wm */ { 0, BX_IA_BSWAP_RX },
bswap undefined behavior
2010-03-19 13:00:48 +03:00
/* 0F C9 /wr */ { 0, BX_IA_BSWAP_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C9 /wm */ { 0, BX_IA_BSWAP_RX },
bswap undefined behavior
2010-03-19 13:00:48 +03:00
/* 0F CA /wr */ { 0, BX_IA_BSWAP_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F CA /wm */ { 0, BX_IA_BSWAP_RX },
bswap undefined behavior
2010-03-19 13:00:48 +03:00
/* 0F CB /wr */ { 0, BX_IA_BSWAP_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F CB /wm */ { 0, BX_IA_BSWAP_RX },
bswap undefined behavior
2010-03-19 13:00:48 +03:00
/* 0F CC /wr */ { 0, BX_IA_BSWAP_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F CC /wm */ { 0, BX_IA_BSWAP_RX },
bswap undefined behavior
2010-03-19 13:00:48 +03:00
/* 0F CD /wr */ { 0, BX_IA_BSWAP_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F CD /wm */ { 0, BX_IA_BSWAP_RX },
bswap undefined behavior
2010-03-19 13:00:48 +03:00
/* 0F CE /wr */ { 0, BX_IA_BSWAP_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F CE /wm */ { 0, BX_IA_BSWAP_RX },
bswap undefined behavior
2010-03-19 13:00:48 +03:00
/* 0F CF /wr */ { 0, BX_IA_BSWAP_RX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F CF /wm */ { 0, BX_IA_BSWAP_RX },
Fixed merge error
2008-08-30 00:43:05 +04:00
/* 0F D0 /wr */ { BxPrefixSSE, BX_IA_ERROR, BxOpcodeGroupSSE_0fd0 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D0 /wm */ { BxPrefixSSE, BX_IA_ERROR, BxOpcodeGroupSSE_0fd0 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F D1 /wr */ { BxPrefixSSE, BX_IA_PSRLW_PqQq, BxOpcodeGroupSSE_0fd1 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D1 /wm */ { BxPrefixSSE, BX_IA_PSRLW_PqQq, BxOpcodeGroupSSE_0fd1 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F D2 /wr */ { BxPrefixSSE, BX_IA_PSRLD_PqQq, BxOpcodeGroupSSE_0fd2 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D2 /wm */ { BxPrefixSSE, BX_IA_PSRLD_PqQq, BxOpcodeGroupSSE_0fd2 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F D3 /wr */ { BxPrefixSSE, BX_IA_PSRLQ_PqQq, BxOpcodeGroupSSE_0fd3 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D3 /wm */ { BxPrefixSSE, BX_IA_PSRLQ_PqQq, BxOpcodeGroupSSE_0fd3 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F D4 /wr */ { BxPrefixSSE, BX_IA_PADDQ_PqQq, BxOpcodeGroupSSE_0fd4 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D4 /wm */ { BxPrefixSSE, BX_IA_PADDQ_PqQq, BxOpcodeGroupSSE_0fd4 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F D5 /wr */ { BxPrefixSSE, BX_IA_PMULLW_PqQq, BxOpcodeGroupSSE_0fd5 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D5 /wm */ { BxPrefixSSE, BX_IA_PMULLW_PqQq, BxOpcodeGroupSSE_0fd5 },
split more SSE ops
2009-11-25 23:49:47 +03:00
/* 0F D6 /wr */ { BxPrefixSSE, BX_IA_ERROR, BxOpcodeGroupSSE_0fd6R },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D6 /wm */ { BxPrefixSSE66, BX_IA_MOVQ_WqVqM },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F D7 /wr */ { BxPrefixSSE, BX_IA_PMOVMSKB_GdPRq, BxOpcodeGroupSSE_0fd7 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D7 /wm */ { 0, BX_IA_ERROR }, // PMOVMSKB is reg/reg form only
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F D8 /wr */ { BxPrefixSSE, BX_IA_PSUBUSB_PqQq, BxOpcodeGroupSSE_0fd8 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D8 /wm */ { BxPrefixSSE, BX_IA_PSUBUSB_PqQq, BxOpcodeGroupSSE_0fd8 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F D9 /wr */ { BxPrefixSSE, BX_IA_PSUBUSW_PqQq, BxOpcodeGroupSSE_0fd9 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D9 /wm */ { BxPrefixSSE, BX_IA_PSUBUSW_PqQq, BxOpcodeGroupSSE_0fd9 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F DA /wr */ { BxPrefixSSE, BX_IA_PMINUB_PqQq, BxOpcodeGroupSSE_0fda },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F DA /wm */ { BxPrefixSSE, BX_IA_PMINUB_PqQq, BxOpcodeGroupSSE_0fda },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F DB /wr */ { BxPrefixSSE, BX_IA_PAND_PqQq, BxOpcodeGroupSSE_0fdb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F DB /wm */ { BxPrefixSSE, BX_IA_PAND_PqQq, BxOpcodeGroupSSE_0fdb },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F DC /wr */ { BxPrefixSSE, BX_IA_PADDUSB_PqQq, BxOpcodeGroupSSE_0fdc },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F DC /wm */ { BxPrefixSSE, BX_IA_PADDUSB_PqQq, BxOpcodeGroupSSE_0fdc },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F DD /wr */ { BxPrefixSSE, BX_IA_PADDUSW_PqQq, BxOpcodeGroupSSE_0fdd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F DD /wm */ { BxPrefixSSE, BX_IA_PADDUSW_PqQq, BxOpcodeGroupSSE_0fdd },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F DE /wr */ { BxPrefixSSE, BX_IA_PMAXUB_PqQq, BxOpcodeGroupSSE_0fde },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F DE /wm */ { BxPrefixSSE, BX_IA_PMAXUB_PqQq, BxOpcodeGroupSSE_0fde },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F DF /wr */ { BxPrefixSSE, BX_IA_PANDN_PqQq, BxOpcodeGroupSSE_0fdf },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F DF /wm */ { BxPrefixSSE, BX_IA_PANDN_PqQq, BxOpcodeGroupSSE_0fdf },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F E0 /wr */ { BxPrefixSSE, BX_IA_PAVGB_PqQq, BxOpcodeGroupSSE_0fe0 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E0 /wm */ { BxPrefixSSE, BX_IA_PAVGB_PqQq, BxOpcodeGroupSSE_0fe0 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F E1 /wr */ { BxPrefixSSE, BX_IA_PSRAW_PqQq, BxOpcodeGroupSSE_0fe1 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E1 /wm */ { BxPrefixSSE, BX_IA_PSRAW_PqQq, BxOpcodeGroupSSE_0fe1 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F E2 /wr */ { BxPrefixSSE, BX_IA_PSRAD_PqQq, BxOpcodeGroupSSE_0fe2 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E2 /wm */ { BxPrefixSSE, BX_IA_PSRAD_PqQq, BxOpcodeGroupSSE_0fe2 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F E3 /wr */ { BxPrefixSSE, BX_IA_PAVGW_PqQq, BxOpcodeGroupSSE_0fe3 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E3 /wm */ { BxPrefixSSE, BX_IA_PAVGW_PqQq, BxOpcodeGroupSSE_0fe3 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F E4 /wr */ { BxPrefixSSE, BX_IA_PMULHUW_PqQq, BxOpcodeGroupSSE_0fe4 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E4 /wm */ { BxPrefixSSE, BX_IA_PMULHUW_PqQq, BxOpcodeGroupSSE_0fe4 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F E5 /wr */ { BxPrefixSSE, BX_IA_PMULHW_PqQq, BxOpcodeGroupSSE_0fe5 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E5 /wm */ { BxPrefixSSE, BX_IA_PMULHW_PqQq, BxOpcodeGroupSSE_0fe5 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F E6 /wr */ { BxPrefixSSE, BX_IA_ERROR, BxOpcodeGroupSSE_0fe6 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E6 /wm */ { BxPrefixSSE, BX_IA_ERROR, BxOpcodeGroupSSE_0fe6 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F E7 /wr */ { 0, BX_IA_ERROR }, // MOVNTQ/MOVNTDQ SSE group
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E7 /wm */ { BxPrefixSSE, BX_IA_MOVNTQ_MqPq, BxOpcodeGroupSSE_0fe7M },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F E8 /wr */ { BxPrefixSSE, BX_IA_PSUBSB_PqQq, BxOpcodeGroupSSE_0fe8 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E8 /wm */ { BxPrefixSSE, BX_IA_PSUBSB_PqQq, BxOpcodeGroupSSE_0fe8 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F E9 /wr */ { BxPrefixSSE, BX_IA_PSUBSW_PqQq, BxOpcodeGroupSSE_0fe9 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E9 /wm */ { BxPrefixSSE, BX_IA_PSUBSW_PqQq, BxOpcodeGroupSSE_0fe9 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F EA /wr */ { BxPrefixSSE, BX_IA_PMINSW_PqQq, BxOpcodeGroupSSE_0fea },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F EA /wm */ { BxPrefixSSE, BX_IA_PMINSW_PqQq, BxOpcodeGroupSSE_0fea },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F EB /wr */ { BxPrefixSSE, BX_IA_POR_PqQq, BxOpcodeGroupSSE_0feb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F EB /wm */ { BxPrefixSSE, BX_IA_POR_PqQq, BxOpcodeGroupSSE_0feb },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F EC /wr */ { BxPrefixSSE, BX_IA_PADDSB_PqQq, BxOpcodeGroupSSE_0fec },
/* 0F EC /wm */ { BxPrefixSSE, BX_IA_PADDSB_PqQq, BxOpcodeGroupSSE_0fec },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F ED /wr */ { BxPrefixSSE, BX_IA_PADDSW_PqQq, BxOpcodeGroupSSE_0fed },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F ED /wm */ { BxPrefixSSE, BX_IA_PADDSW_PqQq, BxOpcodeGroupSSE_0fed },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F EE /wr */ { BxPrefixSSE, BX_IA_PMAXSW_PqQq, BxOpcodeGroupSSE_0fee },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F EE /wm */ { BxPrefixSSE, BX_IA_PMAXSW_PqQq, BxOpcodeGroupSSE_0fee },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F EF /wr */ { BxPrefixSSE, BX_IA_PXOR_PqQq, BxOpcodeGroupSSE_0fef },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F EF /wm */ { BxPrefixSSE, BX_IA_PXOR_PqQq, BxOpcodeGroupSSE_0fef },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F0 /wr */ { 0, BX_IA_ERROR }, // LDDQU SSE group
fixes and cleanups in disasm and decoder
2010-02-09 22:44:25 +03:00
/* 0F F0 /wm */ { BxPrefixSSEF2, BX_IA_LDDQU_VdqMdq },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F1 /wr */ { BxPrefixSSE, BX_IA_PSLLW_PqQq, BxOpcodeGroupSSE_0ff1 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F F1 /wm */ { BxPrefixSSE, BX_IA_PSLLW_PqQq, BxOpcodeGroupSSE_0ff1 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F2 /wr */ { BxPrefixSSE, BX_IA_PSLLD_PqQq, BxOpcodeGroupSSE_0ff2 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F F2 /wm */ { BxPrefixSSE, BX_IA_PSLLD_PqQq, BxOpcodeGroupSSE_0ff2 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F3 /wr */ { BxPrefixSSE, BX_IA_PSLLQ_PqQq, BxOpcodeGroupSSE_0ff3 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F F3 /wm */ { BxPrefixSSE, BX_IA_PSLLQ_PqQq, BxOpcodeGroupSSE_0ff3 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F4 /wr */ { BxPrefixSSE, BX_IA_PMULUDQ_PqQq, BxOpcodeGroupSSE_0ff4 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F F4 /wm */ { BxPrefixSSE, BX_IA_PMULUDQ_PqQq, BxOpcodeGroupSSE_0ff4 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F5 /wr */ { BxPrefixSSE, BX_IA_PMADDWD_PqQq, BxOpcodeGroupSSE_0ff5 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F F5 /wm */ { BxPrefixSSE, BX_IA_PMADDWD_PqQq, BxOpcodeGroupSSE_0ff5 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F6 /wr */ { BxPrefixSSE, BX_IA_PSADBW_PqQq, BxOpcodeGroupSSE_0ff6 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F F6 /wm */ { BxPrefixSSE, BX_IA_PSADBW_PqQq, BxOpcodeGroupSSE_0ff6 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F7 /wr */ { BxPrefixSSE, BX_IA_MASKMOVQ_PqPRq, BxOpcodeGroupSSE_0ff7R },
Complete MASKMOVDQU decoding fix
2008-04-16 09:56:55 +04:00
/* 0F F7 /wm */ { 0, BX_IA_ERROR }, // MASKMOV SSE group
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F8 /wr */ { BxPrefixSSE, BX_IA_PSUBB_PqQq, BxOpcodeGroupSSE_0ff8 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F F8 /wm */ { BxPrefixSSE, BX_IA_PSUBB_PqQq, BxOpcodeGroupSSE_0ff8 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F9 /wr */ { BxPrefixSSE, BX_IA_PSUBW_PqQq, BxOpcodeGroupSSE_0ff9 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F F9 /wm */ { BxPrefixSSE, BX_IA_PSUBW_PqQq, BxOpcodeGroupSSE_0ff9 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F FA /wr */ { BxPrefixSSE, BX_IA_PSUBD_PqQq, BxOpcodeGroupSSE_0ffa },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F FA /wm */ { BxPrefixSSE, BX_IA_PSUBD_PqQq, BxOpcodeGroupSSE_0ffa },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F FB /wr */ { BxPrefixSSE, BX_IA_PSUBQ_PqQq, BxOpcodeGroupSSE_0ffb },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F FB /wm */ { BxPrefixSSE, BX_IA_PSUBQ_PqQq, BxOpcodeGroupSSE_0ffb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F FC /wr */ { BxPrefixSSE, BX_IA_PADDB_PqQq, BxOpcodeGroupSSE_0ffc },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F FC /wm */ { BxPrefixSSE, BX_IA_PADDB_PqQq, BxOpcodeGroupSSE_0ffc },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F FD /wr */ { BxPrefixSSE, BX_IA_PADDW_PqQq, BxOpcodeGroupSSE_0ffd },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F FD /wm */ { BxPrefixSSE, BX_IA_PADDW_PqQq, BxOpcodeGroupSSE_0ffd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F FE /wr */ { BxPrefixSSE, BX_IA_PADDD_PqQq, BxOpcodeGroupSSE_0ffe },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F FE /wm */ { BxPrefixSSE, BX_IA_PADDD_PqQq, BxOpcodeGroupSSE_0ffe },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F FF /wr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F FF /wm */ { 0, BX_IA_ERROR },
split11b on opcode tables level - split almost eevery splittable instruction will be continued
2007-11-17 15:44:10 +03:00
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
// 512 (/r) + 512 (/m) entries for 32bit mode
/* 00 /dr */ { BxArithDstRM, BX_IA_ADD_GbEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 00 /dm */ { BxLockable, BX_IA_ADD_EbGbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 01 /dr */ { BxArithDstRM, BX_IA_ADD_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 01 /dm */ { BxLockable, BX_IA_ADD_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 02 /dr */ { 0, BX_IA_ADD_GbEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 02 /dm */ { 0, BX_IA_ADD_GbEbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 03 /dr */ { 0, BX_IA_ADD_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 03 /dm */ { 0, BX_IA_ADD_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 04 /dr */ { BxImmediate_Ib, BX_IA_ADD_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 04 /dm */ { BxImmediate_Ib, BX_IA_ADD_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 05 /dr */ { BxImmediate_Id, BX_IA_ADD_EAXId },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 05 /dm */ { BxImmediate_Id, BX_IA_ADD_EAXId },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 06 /dr */ { 0, BX_IA_PUSH32_ES },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 06 /dm */ { 0, BX_IA_PUSH32_ES },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 07 /dr */ { 0, BX_IA_POP32_ES },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 07 /dm */ { 0, BX_IA_POP32_ES },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 08 /dr */ { BxArithDstRM, BX_IA_OR_GbEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 08 /dm */ { BxLockable, BX_IA_OR_EbGbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 09 /dr */ { BxArithDstRM, BX_IA_OR_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 09 /dm */ { BxLockable, BX_IA_OR_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0A /dr */ { 0, BX_IA_OR_GbEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0A /dm */ { 0, BX_IA_OR_GbEbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0B /dr */ { 0, BX_IA_OR_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0B /dm */ { 0, BX_IA_OR_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0C /dr */ { BxImmediate_Ib, BX_IA_OR_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0C /dm */ { BxImmediate_Ib, BX_IA_OR_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0D /dr */ { BxImmediate_Id, BX_IA_OR_EAXId },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0D /dm */ { BxImmediate_Id, BX_IA_OR_EAXId },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0E /dr */ { 0, BX_IA_PUSH32_CS },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0E /dm */ { 0, BX_IA_PUSH32_CS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F /dr */ { 0, BX_IA_ERROR }, // 2-byte escape
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F /dm */ { 0, BX_IA_ERROR }, // 2-byte escape
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 10 /dr */ { BxArithDstRM, BX_IA_ADC_GbEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 10 /dm */ { BxLockable, BX_IA_ADC_EbGbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 11 /dr */ { BxArithDstRM, BX_IA_ADC_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 11 /dm */ { BxLockable, BX_IA_ADC_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 12 /dr */ { 0, BX_IA_ADC_GbEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 12 /dm */ { 0, BX_IA_ADC_GbEbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 13 /dr */ { 0, BX_IA_ADC_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 13 /dm */ { 0, BX_IA_ADC_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 14 /dr */ { BxImmediate_Ib, BX_IA_ADC_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 14 /dm */ { BxImmediate_Ib, BX_IA_ADC_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 15 /dr */ { BxImmediate_Id, BX_IA_ADC_EAXId },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 15 /dm */ { BxImmediate_Id, BX_IA_ADC_EAXId },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 16 /dr */ { 0, BX_IA_PUSH32_SS },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 16 /dm */ { 0, BX_IA_PUSH32_SS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 17 /dr */ { BxTraceEnd, BX_IA_POP32_SS }, // async_event = 1
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 17 /dm */ { BxTraceEnd, BX_IA_POP32_SS }, // async_event = 1
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 18 /dr */ { BxArithDstRM, BX_IA_SBB_GbEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 18 /dm */ { BxLockable, BX_IA_SBB_EbGbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 19 /dr */ { BxArithDstRM, BX_IA_SBB_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 19 /dm */ { BxLockable, BX_IA_SBB_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 1A /dr */ { 0, BX_IA_SBB_GbEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 1A /dm */ { 0, BX_IA_SBB_GbEbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 1B /dr */ { 0, BX_IA_SBB_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 1B /dm */ { 0, BX_IA_SBB_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 1C /dr */ { BxImmediate_Ib, BX_IA_SBB_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 1C /dm */ { BxImmediate_Ib, BX_IA_SBB_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 1D /dr */ { BxImmediate_Id, BX_IA_SBB_EAXId },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 1D /dm */ { BxImmediate_Id, BX_IA_SBB_EAXId },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 1E /dr */ { 0, BX_IA_PUSH32_DS },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 1E /dm */ { 0, BX_IA_PUSH32_DS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 1F /dr */ { 0, BX_IA_POP32_DS },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 1F /dm */ { 0, BX_IA_POP32_DS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 20 /dr */ { BxArithDstRM, BX_IA_AND_GbEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 20 /dm */ { BxLockable, BX_IA_AND_EbGbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 21 /dr */ { BxArithDstRM, BX_IA_AND_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 21 /dm */ { BxLockable, BX_IA_AND_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 22 /dr */ { 0, BX_IA_AND_GbEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 22 /dm */ { 0, BX_IA_AND_GbEbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 23 /dr */ { 0, BX_IA_AND_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 23 /dm */ { 0, BX_IA_AND_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 24 /dr */ { BxImmediate_Ib, BX_IA_AND_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 24 /dm */ { BxImmediate_Ib, BX_IA_AND_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 25 /dr */ { BxImmediate_Id, BX_IA_AND_EAXId },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 25 /dm */ { BxImmediate_Id, BX_IA_AND_EAXId },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 26 /dr */ { 0, BX_IA_ERROR }, // ES:
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 26 /dm */ { 0, BX_IA_ERROR }, // ES:
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 27 /dr */ { 0, BX_IA_DAA },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 27 /dm */ { 0, BX_IA_DAA },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 28 /dr */ { BxArithDstRM, BX_IA_SUB_GbEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 28 /dm */ { BxLockable, BX_IA_SUB_EbGbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 29 /dr */ { BxArithDstRM, BX_IA_SUB_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 29 /dm */ { BxLockable, BX_IA_SUB_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 2A /dr */ { 0, BX_IA_SUB_GbEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 2A /dm */ { 0, BX_IA_SUB_GbEbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 2B /dr */ { 0, BX_IA_SUB_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 2B /dm */ { 0, BX_IA_SUB_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 2C /dr */ { BxImmediate_Ib, BX_IA_SUB_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 2C /dm */ { BxImmediate_Ib, BX_IA_SUB_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 2D /dr */ { BxImmediate_Id, BX_IA_SUB_EAXId },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 2D /dm */ { BxImmediate_Id, BX_IA_SUB_EAXId },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 2E /dr */ { 0, BX_IA_ERROR }, // CS:
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 2E /dm */ { 0, BX_IA_ERROR }, // CS:
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 2F /dr */ { 0, BX_IA_DAS },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 2F /dm */ { 0, BX_IA_DAS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 30 /dr */ { BxArithDstRM, BX_IA_XOR_GbEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 30 /dm */ { BxLockable, BX_IA_XOR_EbGbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 31 /dr */ { BxArithDstRM, BX_IA_XOR_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 31 /dm */ { BxLockable, BX_IA_XOR_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 32 /dr */ { 0, BX_IA_XOR_GbEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 32 /dm */ { 0, BX_IA_XOR_GbEbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 33 /dr */ { 0, BX_IA_XOR_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 33 /dm */ { 0, BX_IA_XOR_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 34 /dr */ { BxImmediate_Ib, BX_IA_XOR_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 34 /dm */ { BxImmediate_Ib, BX_IA_XOR_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 35 /dr */ { BxImmediate_Id, BX_IA_XOR_EAXId },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 35 /dm */ { BxImmediate_Id, BX_IA_XOR_EAXId },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 36 /dr */ { 0, BX_IA_ERROR }, // SS:
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 36 /dm */ { 0, BX_IA_ERROR }, // SS:
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 37 /dr */ { 0, BX_IA_AAA },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 37 /dm */ { 0, BX_IA_AAA },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 38 /dr */ { BxArithDstRM, BX_IA_CMP_GbEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 38 /dm */ { 0, BX_IA_CMP_EbGbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 39 /dr */ { BxArithDstRM, BX_IA_CMP_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 39 /dm */ { 0, BX_IA_CMP_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 3A /dr */ { 0, BX_IA_CMP_GbEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 3A /dm */ { 0, BX_IA_CMP_GbEbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 3B /dr */ { 0, BX_IA_CMP_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 3B /dm */ { 0, BX_IA_CMP_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 3C /dr */ { BxImmediate_Ib, BX_IA_CMP_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 3C /dm */ { BxImmediate_Ib, BX_IA_CMP_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 3D /dr */ { BxImmediate_Id, BX_IA_CMP_EAXId },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 3D /dm */ { BxImmediate_Id, BX_IA_CMP_EAXId },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 3E /dr */ { 0, BX_IA_ERROR }, // DS:
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 3E /dm */ { 0, BX_IA_ERROR }, // DS:
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 3F /dr */ { 0, BX_IA_AAS },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 3F /dm */ { 0, BX_IA_AAS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 40 /dr */ { 0, BX_IA_INC_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 40 /dm */ { 0, BX_IA_INC_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 41 /dr */ { 0, BX_IA_INC_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 41 /dm */ { 0, BX_IA_INC_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 42 /dr */ { 0, BX_IA_INC_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 42 /dm */ { 0, BX_IA_INC_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 43 /dr */ { 0, BX_IA_INC_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 43 /dm */ { 0, BX_IA_INC_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 44 /dr */ { 0, BX_IA_INC_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 44 /dm */ { 0, BX_IA_INC_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 45 /dr */ { 0, BX_IA_INC_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 45 /dm */ { 0, BX_IA_INC_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 46 /dr */ { 0, BX_IA_INC_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 46 /dm */ { 0, BX_IA_INC_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 47 /dr */ { 0, BX_IA_INC_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 47 /dm */ { 0, BX_IA_INC_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 48 /dr */ { 0, BX_IA_DEC_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 48 /dm */ { 0, BX_IA_DEC_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 49 /dr */ { 0, BX_IA_DEC_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 49 /dm */ { 0, BX_IA_DEC_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 4A /dr */ { 0, BX_IA_DEC_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 4A /dm */ { 0, BX_IA_DEC_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 4B /dr */ { 0, BX_IA_DEC_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 4B /dm */ { 0, BX_IA_DEC_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 4C /dr */ { 0, BX_IA_DEC_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 4C /dm */ { 0, BX_IA_DEC_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 4D /dr */ { 0, BX_IA_DEC_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 4D /dm */ { 0, BX_IA_DEC_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 4E /dr */ { 0, BX_IA_DEC_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 4E /dm */ { 0, BX_IA_DEC_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 4F /dr */ { 0, BX_IA_DEC_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 4F /dm */ { 0, BX_IA_DEC_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 50 /dr */ { 0, BX_IA_PUSH_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 50 /dm */ { 0, BX_IA_PUSH_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 51 /dr */ { 0, BX_IA_PUSH_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 51 /dm */ { 0, BX_IA_PUSH_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 52 /dr */ { 0, BX_IA_PUSH_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 52 /dm */ { 0, BX_IA_PUSH_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 53 /dr */ { 0, BX_IA_PUSH_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 53 /dm */ { 0, BX_IA_PUSH_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 54 /dr */ { 0, BX_IA_PUSH_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 54 /dm */ { 0, BX_IA_PUSH_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 55 /dr */ { 0, BX_IA_PUSH_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 55 /dm */ { 0, BX_IA_PUSH_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 56 /dr */ { 0, BX_IA_PUSH_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 56 /dm */ { 0, BX_IA_PUSH_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 57 /dr */ { 0, BX_IA_PUSH_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 57 /dm */ { 0, BX_IA_PUSH_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 58 /dr */ { 0, BX_IA_POP_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 58 /dm */ { 0, BX_IA_POP_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 59 /dr */ { 0, BX_IA_POP_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 59 /dm */ { 0, BX_IA_POP_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 5A /dr */ { 0, BX_IA_POP_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 5A /dm */ { 0, BX_IA_POP_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 5B /dr */ { 0, BX_IA_POP_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 5B /dm */ { 0, BX_IA_POP_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 5C /dr */ { 0, BX_IA_POP_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 5C /dm */ { 0, BX_IA_POP_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 5D /dr */ { 0, BX_IA_POP_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 5D /dm */ { 0, BX_IA_POP_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 5E /dr */ { 0, BX_IA_POP_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 5E /dm */ { 0, BX_IA_POP_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 5F /dr */ { 0, BX_IA_POP_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 5F /dm */ { 0, BX_IA_POP_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 60 /dr */ { 0, BX_IA_PUSHAD32 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 60 /dm */ { 0, BX_IA_PUSHAD32 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 61 /dr */ { 0, BX_IA_POPAD32 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 61 /dm */ { 0, BX_IA_POPAD32 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 62 /dr */ { 0, BX_IA_ERROR }, // BOUND
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 62 /dm */ { 0, BX_IA_BOUND_GdMa },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 63 /dr */ { 0, BX_IA_ARPL_EwGw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 63 /dm */ { 0, BX_IA_ARPL_EwGw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 64 /dr */ { 0, BX_IA_ERROR }, // FS:
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 64 /dm */ { 0, BX_IA_ERROR }, // FS:
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 65 /dr */ { 0, BX_IA_ERROR }, // GS:
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 65 /dm */ { 0, BX_IA_ERROR }, // GS:
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 66 /dr */ { 0, BX_IA_ERROR }, // OS:
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 66 /dm */ { 0, BX_IA_ERROR }, // OS:
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 67 /dr */ { 0, BX_IA_ERROR }, // AS:
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 67 /dm */ { 0, BX_IA_ERROR }, // AS:
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 68 /dr */ { BxImmediate_Id, BX_IA_PUSH_Id },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 68 /dm */ { BxImmediate_Id, BX_IA_PUSH_Id },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 69 /dr */ { BxImmediate_Id, BX_IA_IMUL_GdEdIdR },
Split more opcodes using new LOAD technique
2008-08-11 01:16:12 +04:00
/* 69 /dm */ { BxImmediate_Id, BX_IA_IMUL_GdEdIdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 6A /dr */ { BxImmediate_Ib_SE, BX_IA_PUSH_Id },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 6A /dm */ { BxImmediate_Ib_SE, BX_IA_PUSH_Id },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 6B /dr */ { BxImmediate_Ib_SE, BX_IA_IMUL_GdEdIdR },
Split more opcodes using new LOAD technique
2008-08-11 01:16:12 +04:00
/* 6B /dm */ { BxImmediate_Ib_SE, BX_IA_IMUL_GdEdIdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 6C /dr */ { 0, BX_IA_REP_INSB_YbDX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 6C /dm */ { 0, BX_IA_REP_INSB_YbDX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 6D /dr */ { 0, BX_IA_REP_INSD_YdDX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 6D /dm */ { 0, BX_IA_REP_INSD_YdDX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 6E /dr */ { 0, BX_IA_REP_OUTSB_DXXb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 6E /dm */ { 0, BX_IA_REP_OUTSB_DXXb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 6F /dr */ { 0, BX_IA_REP_OUTSD_DXXd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 6F /dm */ { 0, BX_IA_REP_OUTSD_DXXd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 70 /dr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JO_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 70 /dm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JO_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 71 /dr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNO_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 71 /dm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNO_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 72 /dr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JB_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 72 /dm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JB_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 73 /dr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNB_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 73 /dm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNB_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 74 /dr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JZ_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 74 /dm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JZ_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 75 /dr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNZ_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 75 /dm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNZ_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 76 /dr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JBE_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 76 /dm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JBE_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 77 /dr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNBE_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 77 /dm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNBE_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 78 /dr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JS_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 78 /dm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JS_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 79 /dr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNS_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 79 /dm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNS_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 7A /dr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JP_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 7A /dm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JP_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 7B /dr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNP_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 7B /dm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNP_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 7C /dr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JL_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 7C /dm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JL_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 7D /dr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNL_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 7D /dm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNL_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 7E /dr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JLE_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 7E /dm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JLE_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 7F /dr */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNLE_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 7F /dm */ { BxImmediate_BrOff8 | BxTraceJCC, BX_IA_JNLE_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 80 /dr */ { BxGroup1 | BxImmediate_Ib, BX_IA_ERROR, BxOpcodeInfoG1EbIbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 80 /dm */ { BxGroup1 | BxImmediate_Ib, BX_IA_ERROR, BxOpcodeInfoG1EbIbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 81 /dr */ { BxGroup1 | BxImmediate_Id, BX_IA_ERROR, BxOpcodeInfoG1EdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 81 /dm */ { BxGroup1 | BxImmediate_Id, BX_IA_ERROR, BxOpcodeInfoG1EdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 82 /dr */ { BxGroup1 | BxImmediate_Ib, BX_IA_ERROR, BxOpcodeInfoG1EbIbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 82 /dm */ { BxGroup1 | BxImmediate_Ib, BX_IA_ERROR, BxOpcodeInfoG1EbIbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 83 /dr */ { BxGroup1 | BxImmediate_Ib_SE, BX_IA_ERROR, BxOpcodeInfoG1EdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 83 /dm */ { BxGroup1 | BxImmediate_Ib_SE, BX_IA_ERROR, BxOpcodeInfoG1EdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 84 /dr */ { 0, BX_IA_TEST_EbGbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 84 /dm */ { 0, BX_IA_TEST_EbGbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 85 /dr */ { 0, BX_IA_TEST_EdGdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 85 /dm */ { 0, BX_IA_TEST_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 86 /dr */ { 0, BX_IA_XCHG_EbGbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 86 /dm */ { BxLockable, BX_IA_XCHG_EbGbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 87 /dr */ { 0, BX_IA_XCHG_EdGdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 87 /dm */ { BxLockable, BX_IA_XCHG_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 88 /dr */ { BxArithDstRM, BX_IA_MOV_GbEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 88 /dm */ { 0, BX_IA_MOV_EbGbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 89 /dr */ { BxArithDstRM, BX_IA_MOV_GdEdR },
Split instruction 32/64
2009-04-06 22:44:28 +04:00
/* 89 /dm */ { 0, BX_IA_MOV32_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 8A /dr */ { 0, BX_IA_MOV_GbEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 8A /dm */ { 0, BX_IA_MOV_GbEbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 8B /dr */ { 0, BX_IA_MOV_GdEdR },
small non-significant speedups
2008-09-07 01:10:40 +04:00
/* 8B /dm */ { 0, BX_IA_MOV32_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 8C /dr */ { 0, BX_IA_MOV_EwSwR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 8C /dm */ { 0, BX_IA_MOV_EwSwM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 8D /dr */ { 0, BX_IA_ERROR }, // LEA
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 8D /dm */ { 0, BX_IA_LEA_GdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 8E /dr */ { 0, BX_IA_MOV_SwEw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 8E /dm */ { 0, BX_IA_MOV_SwEw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 8F /dr */ { BxGroup1A, BX_IA_ERROR, BxOpcodeInfoG1AEdR },
More fetchdecode fixes
2008-05-01 01:32:33 +04:00
/* 8F /dm */ { BxGroup1A, BX_IA_ERROR, BxOpcodeInfoG1AEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 90 /dr */ { BxPrefixSSE, BX_IA_NOP, BxOpcodeGroupSSE_PAUSE },
Separate PAUSE instruction from regular NOP
2009-01-27 23:29:05 +03:00
/* 90 /dm */ { BxPrefixSSE, BX_IA_NOP, BxOpcodeGroupSSE_PAUSE },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 91 /dr */ { 0, BX_IA_XCHG_ERXEAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 91 /dm */ { 0, BX_IA_XCHG_ERXEAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 92 /dr */ { 0, BX_IA_XCHG_ERXEAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 92 /dm */ { 0, BX_IA_XCHG_ERXEAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 93 /dr */ { 0, BX_IA_XCHG_ERXEAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 93 /dm */ { 0, BX_IA_XCHG_ERXEAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 94 /dr */ { 0, BX_IA_XCHG_ERXEAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 94 /dm */ { 0, BX_IA_XCHG_ERXEAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 95 /dr */ { 0, BX_IA_XCHG_ERXEAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 95 /dm */ { 0, BX_IA_XCHG_ERXEAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 96 /dr */ { 0, BX_IA_XCHG_ERXEAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 96 /dm */ { 0, BX_IA_XCHG_ERXEAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 97 /dr */ { 0, BX_IA_XCHG_ERXEAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 97 /dm */ { 0, BX_IA_XCHG_ERXEAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 98 /dr */ { 0, BX_IA_CWDE },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 98 /dm */ { 0, BX_IA_CWDE },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 99 /dr */ { 0, BX_IA_CDQ },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 99 /dm */ { 0, BX_IA_CDQ },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 9A /dr */ { BxImmediate_IdIw | BxTraceEnd, BX_IA_CALL32_Ap },
branch_16 optimizations
2008-08-24 02:27:58 +04:00
/* 9A /dm */ { BxImmediate_IdIw | BxTraceEnd, BX_IA_CALL32_Ap },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 9B /dr */ { 0, BX_IA_FWAIT },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 9B /dm */ { 0, BX_IA_FWAIT },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 9C /dr */ { 0, BX_IA_PUSHF_Fd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 9C /dm */ { 0, BX_IA_PUSHF_Fd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 9D /dr */ { 0, BX_IA_POPF_Fd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 9D /dm */ { 0, BX_IA_POPF_Fd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 9E /dr */ { 0, BX_IA_SAHF },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 9E /dm */ { 0, BX_IA_SAHF },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 9F /dr */ { 0, BX_IA_LAHF },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 9F /dm */ { 0, BX_IA_LAHF },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A0 /dr */ { BxImmediate_O, BX_IA_MOV_ALOd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A0 /dm */ { BxImmediate_O, BX_IA_MOV_ALOd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A1 /dr */ { BxImmediate_O, BX_IA_MOV_EAXOd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A1 /dm */ { BxImmediate_O, BX_IA_MOV_EAXOd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A2 /dr */ { BxImmediate_O, BX_IA_MOV_OdAL },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A2 /dm */ { BxImmediate_O, BX_IA_MOV_OdAL },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A3 /dr */ { BxImmediate_O, BX_IA_MOV_OdEAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A3 /dm */ { BxImmediate_O, BX_IA_MOV_OdEAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A4 /dr */ { 0, BX_IA_REP_MOVSB_XbYb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A4 /dm */ { 0, BX_IA_REP_MOVSB_XbYb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A5 /dr */ { 0, BX_IA_REP_MOVSD_XdYd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A5 /dm */ { 0, BX_IA_REP_MOVSD_XdYd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A6 /dr */ { 0, BX_IA_REP_CMPSB_XbYb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A6 /dm */ { 0, BX_IA_REP_CMPSB_XbYb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A7 /dr */ { 0, BX_IA_REP_CMPSD_XdYd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A7 /dm */ { 0, BX_IA_REP_CMPSD_XdYd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A8 /dr */ { BxImmediate_Ib, BX_IA_TEST_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A8 /dm */ { BxImmediate_Ib, BX_IA_TEST_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* A9 /dr */ { BxImmediate_Id, BX_IA_TEST_EAXId },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* A9 /dm */ { BxImmediate_Id, BX_IA_TEST_EAXId },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* AA /dr */ { 0, BX_IA_REP_STOSB_YbAL },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* AA /dm */ { 0, BX_IA_REP_STOSB_YbAL },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* AB /dr */ { 0, BX_IA_REP_STOSD_YdEAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* AB /dm */ { 0, BX_IA_REP_STOSD_YdEAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* AC /dr */ { 0, BX_IA_REP_LODSB_ALXb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* AC /dm */ { 0, BX_IA_REP_LODSB_ALXb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* AD /dr */ { 0, BX_IA_REP_LODSD_EAXXd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* AD /dm */ { 0, BX_IA_REP_LODSD_EAXXd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* AE /dr */ { 0, BX_IA_REP_SCASB_ALXb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* AE /dm */ { 0, BX_IA_REP_SCASB_ALXb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* AF /dr */ { 0, BX_IA_REP_SCASD_EAXXd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* AF /dm */ { 0, BX_IA_REP_SCASD_EAXXd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B0 /dr */ { BxImmediate_Ib, BX_IA_MOV_RLIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B0 /dm */ { BxImmediate_Ib, BX_IA_MOV_RLIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B1 /dr */ { BxImmediate_Ib, BX_IA_MOV_RLIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B1 /dm */ { BxImmediate_Ib, BX_IA_MOV_RLIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B2 /dr */ { BxImmediate_Ib, BX_IA_MOV_RLIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B2 /dm */ { BxImmediate_Ib, BX_IA_MOV_RLIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B3 /dr */ { BxImmediate_Ib, BX_IA_MOV_RLIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B3 /dm */ { BxImmediate_Ib, BX_IA_MOV_RLIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B4 /dr */ { BxImmediate_Ib, BX_IA_MOV_RHIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B4 /dm */ { BxImmediate_Ib, BX_IA_MOV_RHIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B5 /dr */ { BxImmediate_Ib, BX_IA_MOV_RHIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B5 /dm */ { BxImmediate_Ib, BX_IA_MOV_RHIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B6 /dr */ { BxImmediate_Ib, BX_IA_MOV_RHIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B6 /dm */ { BxImmediate_Ib, BX_IA_MOV_RHIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B7 /dr */ { BxImmediate_Ib, BX_IA_MOV_RHIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B7 /dm */ { BxImmediate_Ib, BX_IA_MOV_RHIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B8 /dr */ { BxImmediate_Id, BX_IA_MOV_ERXId },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B8 /dm */ { BxImmediate_Id, BX_IA_MOV_ERXId },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* B9 /dr */ { BxImmediate_Id, BX_IA_MOV_ERXId },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* B9 /dm */ { BxImmediate_Id, BX_IA_MOV_ERXId },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* BA /dr */ { BxImmediate_Id, BX_IA_MOV_ERXId },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* BA /dm */ { BxImmediate_Id, BX_IA_MOV_ERXId },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* BB /dr */ { BxImmediate_Id, BX_IA_MOV_ERXId },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* BB /dm */ { BxImmediate_Id, BX_IA_MOV_ERXId },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* BC /dr */ { BxImmediate_Id, BX_IA_MOV_ERXId },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* BC /dm */ { BxImmediate_Id, BX_IA_MOV_ERXId },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* BD /dr */ { BxImmediate_Id, BX_IA_MOV_ERXId },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* BD /dm */ { BxImmediate_Id, BX_IA_MOV_ERXId },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* BE /dr */ { BxImmediate_Id, BX_IA_MOV_ERXId },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* BE /dm */ { BxImmediate_Id, BX_IA_MOV_ERXId },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* BF /dr */ { BxImmediate_Id, BX_IA_MOV_ERXId },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* BF /dm */ { BxImmediate_Id, BX_IA_MOV_ERXId },
Split shift opcodes | optimize SAR opcode
2010-05-18 11:28:05 +04:00
/* C0 /dr */ { BxGroup2 | BxImmediate_Ib, BX_IA_ERROR, BxOpcodeInfoG2EbR },
/* C0 /dm */ { BxGroup2 | BxImmediate_Ib, BX_IA_ERROR, BxOpcodeInfoG2EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* C1 /dr */ { BxGroup2 | BxImmediate_Ib, BX_IA_ERROR, BxOpcodeInfoG2EdR },
Split shift/rotate opcodes in 32-bit mode and 64-bit mode
2008-05-03 02:47:07 +04:00
/* C1 /dm */ { BxGroup2 | BxImmediate_Ib, BX_IA_ERROR, BxOpcodeInfoG2EdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* C2 /dr */ { BxImmediate_Iw | BxTraceEnd, BX_IA_RETnear32_Iw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* C2 /dm */ { BxImmediate_Iw | BxTraceEnd, BX_IA_RETnear32_Iw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* C3 /dr */ { BxTraceEnd, BX_IA_RETnear32 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* C3 /dm */ { BxTraceEnd, BX_IA_RETnear32 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* C4 /dr */ { 0, BX_IA_ERROR }, // LES
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* C4 /dm */ { 0, BX_IA_LES_GdMp },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* C5 /dr */ { 0, BX_IA_ERROR }, // LDS
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* C5 /dm */ { 0, BX_IA_LDS_GdMp },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* C6 /dr */ { BxGroup11, BX_IA_ERROR, BxOpcodeInfoG11EbR },
Correctly implement opcode group G11 i.e. instructions C6 and C7 should @UD when modrm nnn field != 0 (1st instr in the group
2008-04-25 01:52:28 +04:00
/* C6 /dm */ { BxGroup11, BX_IA_ERROR, BxOpcodeInfoG11EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* C7 /dr */ { BxGroup11, BX_IA_ERROR, BxOpcodeInfoG11EdR },
Correctly implement opcode group G11 i.e. instructions C6 and C7 should @UD when modrm nnn field != 0 (1st instr in the group
2008-04-25 01:52:28 +04:00
/* C7 /dm */ { BxGroup11, BX_IA_ERROR, BxOpcodeInfoG11EdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* C8 /dr */ { BxImmediate_IwIb, BX_IA_ENTER32_IwIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* C8 /dm */ { BxImmediate_IwIb, BX_IA_ENTER32_IwIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* C9 /dr */ { 0, BX_IA_LEAVE32 },
optimize LEAVE operation
2008-08-28 01:57:40 +04:00
/* C9 /dm */ { 0, BX_IA_LEAVE32 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* CA /dr */ { BxImmediate_Iw | BxTraceEnd, BX_IA_RETfar32_Iw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* CA /dm */ { BxImmediate_Iw | BxTraceEnd, BX_IA_RETfar32_Iw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* CB /dr */ { BxTraceEnd, BX_IA_RETfar32 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* CB /dm */ { BxTraceEnd, BX_IA_RETfar32 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* CC /dr */ { BxTraceEnd, BX_IA_INT3 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* CC /dm */ { BxTraceEnd, BX_IA_INT3 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* CD /dr */ { BxImmediate_Ib | BxTraceEnd, BX_IA_INT_Ib },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* CD /dm */ { BxImmediate_Ib | BxTraceEnd, BX_IA_INT_Ib },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* CE /dr */ { BxTraceEnd, BX_IA_INTO },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* CE /dm */ { BxTraceEnd, BX_IA_INTO },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* CF /dr */ { BxTraceEnd, BX_IA_IRET32 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* CF /dm */ { BxTraceEnd, BX_IA_IRET32 },
Split shift opcodes | optimize SAR opcode
2010-05-18 11:28:05 +04:00
/* D0 /dr */ { BxGroup2 | BxImmediate_I1, BX_IA_ERROR, BxOpcodeInfoG2EbR },
/* D0 /dm */ { BxGroup2 | BxImmediate_I1, BX_IA_ERROR, BxOpcodeInfoG2EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* D1 /dr */ { BxGroup2 | BxImmediate_I1, BX_IA_ERROR, BxOpcodeInfoG2EdR },
Split shift/rotate opcodes in 32-bit mode and 64-bit mode
2008-05-03 02:47:07 +04:00
/* D1 /dm */ { BxGroup2 | BxImmediate_I1, BX_IA_ERROR, BxOpcodeInfoG2EdM },
Split shift opcodes | optimize SAR opcode
2010-05-18 11:28:05 +04:00
/* D2 /dr */ { BxGroup2, BX_IA_ERROR, BxOpcodeInfoG2EbR },
/* D2 /dm */ { BxGroup2, BX_IA_ERROR, BxOpcodeInfoG2EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* D3 /dr */ { BxGroup2, BX_IA_ERROR, BxOpcodeInfoG2EdR },
Split shift/rotate opcodes in 32-bit mode and 64-bit mode
2008-05-03 02:47:07 +04:00
/* D3 /dm */ { BxGroup2, BX_IA_ERROR, BxOpcodeInfoG2EdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* D4 /dr */ { BxImmediate_Ib, BX_IA_AAM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* D4 /dm */ { BxImmediate_Ib, BX_IA_AAM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* D5 /dr */ { BxImmediate_Ib, BX_IA_AAD },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* D5 /dm */ { BxImmediate_Ib, BX_IA_AAD },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* D6 /dr */ { 0, BX_IA_SALC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* D6 /dm */ { 0, BX_IA_SALC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* D7 /dr */ { 0, BX_IA_XLAT },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* D7 /dm */ { 0, BX_IA_XLAT },
split11b on opcode tables level - split almost eevery splittable instruction will be continued
2007-11-17 15:44:10 +03:00
#if BX_SUPPORT_FPU
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* D8 /dr */ { BxGroupN, BX_IA_ERROR, BxOpcodeInfo_FloatingPointD8 },
A bit simplify FPU decoding
2008-09-16 22:28:53 +04:00
/* D8 /dm */ { BxGroupFP, BX_IA_ERROR, BxOpcodeInfo_FPGroupD8 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* D9 /dr */ { BxFPEscape, BX_IA_ERROR, BxOpcodeInfo_FloatingPointD9 },
A bit simplify FPU decoding
2008-09-16 22:28:53 +04:00
/* D9 /dm */ { BxGroupFP, BX_IA_ERROR, BxOpcodeInfo_FPGroupD9 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DA /dr */ { BxFPEscape, BX_IA_ERROR, BxOpcodeInfo_FloatingPointDA },
A bit simplify FPU decoding
2008-09-16 22:28:53 +04:00
/* DA /dm */ { BxGroupFP, BX_IA_ERROR, BxOpcodeInfo_FPGroupDA },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DB /dr */ { BxFPEscape, BX_IA_ERROR, BxOpcodeInfo_FloatingPointDB },
A bit simplify FPU decoding
2008-09-16 22:28:53 +04:00
/* DB /dm */ { BxGroupFP, BX_IA_ERROR, BxOpcodeInfo_FPGroupDB },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DC /dr */ { BxGroupN, BX_IA_ERROR, BxOpcodeInfo_FloatingPointDC },
A bit simplify FPU decoding
2008-09-16 22:28:53 +04:00
/* DC /dm */ { BxGroupFP, BX_IA_ERROR, BxOpcodeInfo_FPGroupDC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DD /dr */ { BxGroupN, BX_IA_ERROR, BxOpcodeInfo_FloatingPointDD },
A bit simplify FPU decoding
2008-09-16 22:28:53 +04:00
/* DD /dm */ { BxGroupFP, BX_IA_ERROR, BxOpcodeInfo_FPGroupDD },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DE /dr */ { BxFPEscape, BX_IA_ERROR, BxOpcodeInfo_FloatingPointDE },
A bit simplify FPU decoding
2008-09-16 22:28:53 +04:00
/* DE /dm */ { BxGroupFP, BX_IA_ERROR, BxOpcodeInfo_FPGroupDE },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DF /dr */ { BxFPEscape, BX_IA_ERROR, BxOpcodeInfo_FloatingPointDF },
A bit simplify FPU decoding
2008-09-16 22:28:53 +04:00
/* DF /dm */ { BxGroupFP, BX_IA_ERROR, BxOpcodeInfo_FPGroupDF },
split11b on opcode tables level - split almost eevery splittable instruction will be continued
2007-11-17 15:44:10 +03:00
#else
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* D8 /dr */ { 0, BX_IA_FPU_ESC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* D8 /dm */ { 0, BX_IA_FPU_ESC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* D9 /dr */ { 0, BX_IA_FPU_ESC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* D9 /dm */ { 0, BX_IA_FPU_ESC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DA /dr */ { 0, BX_IA_FPU_ESC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* DA /dm */ { 0, BX_IA_FPU_ESC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DB /dr */ { 0, BX_IA_FPU_ESC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* DB /dm */ { 0, BX_IA_FPU_ESC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DC /dr */ { 0, BX_IA_FPU_ESC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* DC /dm */ { 0, BX_IA_FPU_ESC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DD /dr */ { 0, BX_IA_FPU_ESC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* DD /dm */ { 0, BX_IA_FPU_ESC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DE /dr */ { 0, BX_IA_FPU_ESC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* DE /dm */ { 0, BX_IA_FPU_ESC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* DF /dr */ { 0, BX_IA_FPU_ESC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* DF /dm */ { 0, BX_IA_FPU_ESC },
split11b on opcode tables level - split almost eevery splittable instruction will be continued
2007-11-17 15:44:10 +03:00
#endif
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E0 /dr */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_LOOPNE32_Jb },
Reorganize ctrl_xfer8.cc code, allows to inline branch32 method
2008-06-22 07:45:55 +04:00
/* E0 /dm */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_LOOPNE32_Jb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E1 /dr */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_LOOPE32_Jb },
Reorganize ctrl_xfer8.cc code, allows to inline branch32 method
2008-06-22 07:45:55 +04:00
/* E1 /dm */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_LOOPE32_Jb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E2 /dr */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_LOOP32_Jb },
Reorganize ctrl_xfer8.cc code, allows to inline branch32 method
2008-06-22 07:45:55 +04:00
/* E2 /dm */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_LOOP32_Jb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E3 /dr */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_JECXZ_Jb },
Reorganize ctrl_xfer8.cc code, allows to inline branch32 method
2008-06-22 07:45:55 +04:00
/* E3 /dm */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_JECXZ_Jb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E4 /dr */ { BxImmediate_Ib, BX_IA_IN_ALIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* E4 /dm */ { BxImmediate_Ib, BX_IA_IN_ALIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E5 /dr */ { BxImmediate_Ib, BX_IA_IN_EAXIb },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* E5 /dm */ { BxImmediate_Ib, BX_IA_IN_EAXIb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E6 /dr */ { BxImmediate_Ib, BX_IA_OUT_IbAL },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* E6 /dm */ { BxImmediate_Ib, BX_IA_OUT_IbAL },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E7 /dr */ { BxImmediate_Ib, BX_IA_OUT_IbEAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* E7 /dm */ { BxImmediate_Ib, BX_IA_OUT_IbEAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E8 /dr */ { BxImmediate_BrOff32 | BxTraceEnd, BX_IA_CALL_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* E8 /dm */ { BxImmediate_BrOff32 | BxTraceEnd, BX_IA_CALL_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* E9 /dr */ { BxImmediate_BrOff32 | BxTraceEnd, BX_IA_JMP_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* E9 /dm */ { BxImmediate_BrOff32 | BxTraceEnd, BX_IA_JMP_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* EA /dr */ { BxImmediate_IdIw | BxTraceEnd, BX_IA_JMP_Ap },
branch_16 optimizations
2008-08-24 02:27:58 +04:00
/* EA /dm */ { BxImmediate_IdIw | BxTraceEnd, BX_IA_JMP_Ap },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* EB /dr */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_JMP_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* EB /dm */ { BxImmediate_BrOff8 | BxTraceEnd, BX_IA_JMP_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* EC /dr */ { 0, BX_IA_IN_ALDX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* EC /dm */ { 0, BX_IA_IN_ALDX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* ED /dr */ { 0, BX_IA_IN_EAXDX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* ED /dm */ { 0, BX_IA_IN_EAXDX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* EE /dr */ { 0, BX_IA_OUT_DXAL },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* EE /dm */ { 0, BX_IA_OUT_DXAL },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* EF /dr */ { 0, BX_IA_OUT_DXEAX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* EF /dm */ { 0, BX_IA_OUT_DXEAX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F0 /dr */ { 0, BX_IA_ERROR }, // LOCK:
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F0 /dm */ { 0, BX_IA_ERROR }, // LOCK:
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F1 /dr */ { BxTraceEnd, BX_IA_INT1 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F1 /dm */ { BxTraceEnd, BX_IA_INT1 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F2 /dr */ { 0, BX_IA_ERROR }, // REPNE/REPNZ
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F2 /dm */ { 0, BX_IA_ERROR }, // REPNE/REPNZ
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F3 /dr */ { 0, BX_IA_ERROR }, // REP,REPE/REPZ
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F3 /dm */ { 0, BX_IA_ERROR }, // REP,REPE/REPZ
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F4 /dr */ { BxTraceEnd, BX_IA_HLT },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F4 /dm */ { BxTraceEnd, BX_IA_HLT },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F5 /dr */ { 0, BX_IA_CMC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F5 /dm */ { 0, BX_IA_CMC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F6 /dr */ { BxGroup3, BX_IA_ERROR, BxOpcodeInfoG3EbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F6 /dm */ { BxGroup3, BX_IA_ERROR, BxOpcodeInfoG3EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F7 /dr */ { BxGroup3, BX_IA_ERROR, BxOpcodeInfoG3EdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F7 /dm */ { BxGroup3, BX_IA_ERROR, BxOpcodeInfoG3EdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F8 /dr */ { 0, BX_IA_CLC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F8 /dm */ { 0, BX_IA_CLC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* F9 /dr */ { 0, BX_IA_STC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* F9 /dm */ { 0, BX_IA_STC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* FA /dr */ { 0, BX_IA_CLI },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* FA /dm */ { 0, BX_IA_CLI },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* FB /dr */ { 0, BX_IA_STI },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* FB /dm */ { 0, BX_IA_STI },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* FC /dr */ { 0, BX_IA_CLD },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* FC /dm */ { 0, BX_IA_CLD },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* FD /dr */ { 0, BX_IA_STD },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* FD /dm */ { 0, BX_IA_STD },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* FE /dr */ { BxGroup5, BX_IA_ERROR, BxOpcodeInfoG4R },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* FE /dm */ { BxGroup4, BX_IA_ERROR, BxOpcodeInfoG4M },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* FF /dr */ { BxGroup5, BX_IA_ERROR, BxOpcodeInfoG5dR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* FF /dm */ { BxGroup5, BX_IA_ERROR, BxOpcodeInfoG5dM },
split11b on opcode tables level - split almost eevery splittable instruction will be continued
2007-11-17 15:44:10 +03:00
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 00 /dr */ { BxGroup6, BX_IA_ERROR, BxOpcodeInfoG6 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 00 /dm */ { BxGroup6, BX_IA_ERROR, BxOpcodeInfoG6 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 01 /dr */ { BxFPEscape, BX_IA_ERROR, BxOpcodeInfoG7R },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 01 /dm */ { BxGroup7, BX_IA_ERROR, BxOpcodeInfoG7M },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 02 /dr */ { 0, BX_IA_LAR_GvEw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 02 /dm */ { 0, BX_IA_LAR_GvEw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 03 /dr */ { 0, BX_IA_LSL_GvEw },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 03 /dm */ { 0, BX_IA_LSL_GvEw },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 04 /dr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 04 /dm */ { 0, BX_IA_ERROR },
split11b on opcode tables level - split almost eevery splittable instruction will be continued
2007-11-17 15:44:10 +03:00
#if BX_SUPPORT_X86_64
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 05 /dr */ { BxTraceEnd, BX_IA_SYSCALL },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 05 /dm */ { BxTraceEnd, BX_IA_SYSCALL },
split11b on opcode tables level - split almost eevery splittable instruction will be continued
2007-11-17 15:44:10 +03:00
#elif BX_CPU_LEVEL == 2
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 05 /dr */ { BxTraceEnd, BX_IA_LOADALL },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 05 /dm */ { BxTraceEnd, BX_IA_LOADALL },
split11b on opcode tables level - split almost eevery splittable instruction will be continued
2007-11-17 15:44:10 +03:00
#else
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 05 /dr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 05 /dm */ { 0, BX_IA_ERROR },
split11b on opcode tables level - split almost eevery splittable instruction will be continued
2007-11-17 15:44:10 +03:00
#endif
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 06 /dr */ { 0, BX_IA_CLTS },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 06 /dm */ { 0, BX_IA_CLTS },
split11b on opcode tables level - split almost eevery splittable instruction will be continued
2007-11-17 15:44:10 +03:00
#if BX_SUPPORT_X86_64
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 07 /dr */ { BxTraceEnd, BX_IA_SYSRET },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 07 /dm */ { BxTraceEnd, BX_IA_SYSRET },
split11b on opcode tables level - split almost eevery splittable instruction will be continued
2007-11-17 15:44:10 +03:00
#else
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 07 /dr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 07 /dm */ { 0, BX_IA_ERROR },
split11b on opcode tables level - split almost eevery splittable instruction will be continued
2007-11-17 15:44:10 +03:00
#endif
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 08 /dr */ { BxTraceEnd, BX_IA_INVD },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 08 /dm */ { BxTraceEnd, BX_IA_INVD },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 09 /dr */ { BxTraceEnd, BX_IA_WBINVD },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 09 /dm */ { BxTraceEnd, BX_IA_WBINVD },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 0A /dr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 0A /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 0B /dr */ { BxTraceEnd, BX_IA_UD2A },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 0B /dm */ { BxTraceEnd, BX_IA_UD2A },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 0C /dr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 0C /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 0D /dr */ { 0, BX_IA_PREFETCHW }, // 3DNow! PREFETCH on AMD, NOP on Intel
1st step of moving CPU options to runtime
2010-02-24 22:27:51 +03:00
/* 0F 0D /dm */ { 0, BX_IA_PREFETCHW }, // 3DNow! PREFETCH on AMD, NOP on Intel
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 0E /dr */ { 0, BX_IA_FEMMS }, // 3DNow! FEMMS
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 0E /dm */ { 0, BX_IA_FEMMS }, // 3DNow! FEMMS
split11b on opcode tables level - split almost eevery splittable instruction will be continued
2007-11-17 15:44:10 +03:00
#if BX_SUPPORT_3DNOW
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 0F /dr */ { BxImmediate_Ib, BX_IA_ERROR },
compilation fix
2010-04-30 01:04:23 +04:00
/* 0F 0F /dm */ { BxImmediate_Ib, BX_IA_ERROR },
split11b on opcode tables level - split almost eevery splittable instruction will be continued
2007-11-17 15:44:10 +03:00
#else
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 0F /dr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 0F /dm */ { 0, BX_IA_ERROR },
split11b on opcode tables level - split almost eevery splittable instruction will be continued
2007-11-17 15:44:10 +03:00
#endif
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 10 /dr */ { BxPrefixSSE, BX_IA_MOVUPS_VpsWpsR, BxOpcodeGroupSSE_0f10R },
split SSE move instructions
2009-10-24 15:17:51 +04:00
/* 0F 10 /dm */ { BxPrefixSSE, BX_IA_MOVUPS_VpsWpsM, BxOpcodeGroupSSE_0f10M },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 11 /dr */ { BxPrefixSSE, BX_IA_MOVUPS_WpsVpsR, BxOpcodeGroupSSE_0f11R },
split SSE move instructions
2009-10-24 15:17:51 +04:00
/* 0F 11 /dm */ { BxPrefixSSE, BX_IA_MOVUPS_WpsVpsM, BxOpcodeGroupSSE_0f11M },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 12 /dr */ { BxPrefixSSE, BX_IA_MOVHLPS_VpsWpsR, BxOpcodeGroupSSE_0f12R },
fix
2010-03-31 18:03:07 +04:00
/* 0F 12 /dm */ { BxPrefixSSE, BX_IA_MOVLPS_VpsMq, BxOpcodeGroupSSE_0f12M },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 13 /dr */ { 0, BX_IA_ERROR }, // MOVLPS/PD SSE group
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 13 /dm */ { BxPrefixSSE, BX_IA_MOVLPS_MqVps, BxOpcodeGroupSSE_0f13M },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 14 /dr */ { BxPrefixSSE, BX_IA_UNPCKLPS_VpsWdq, BxOpcodeGroupSSE_0f14 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 14 /dm */ { BxPrefixSSE, BX_IA_UNPCKLPS_VpsWdq, BxOpcodeGroupSSE_0f14 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 15 /dr */ { BxPrefixSSE, BX_IA_UNPCKHPS_VpsWdq, BxOpcodeGroupSSE_0f15 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 15 /dm */ { BxPrefixSSE, BX_IA_UNPCKHPS_VpsWdq, BxOpcodeGroupSSE_0f15 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 16 /dr */ { BxPrefixSSE, BX_IA_MOVLHPS_VpsWpsR, BxOpcodeGroupSSE_0f16R },
decode fix
2010-03-30 20:39:57 +04:00
/* 0F 16 /dm */ { BxPrefixSSE, BX_IA_MOVHPS_VpsMq, BxOpcodeGroupSSE_0f16M },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 17 /dr */ { 0, BX_IA_ERROR }, // MOHLPS/PD SSE group
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 17 /dm */ { BxPrefixSSE, BX_IA_MOVHPS_MqVps, BxOpcodeGroupSSE_0f17M },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
#if BX_CPU_LEVEL >= 6
/* 0F 18 /dr */ { 0, BX_IA_NOP }, // PREFETCH HINT
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 18 /dm */ { 0, BX_IA_PREFETCH }, // opcode group G16, PREFETCH hints
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 19 /dr */ { 0, BX_IA_NOP }, // multi-byte NOP
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 19 /dm */ { 0, BX_IA_NOP }, // multi-byte NOP
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1A /dr */ { 0, BX_IA_NOP }, // multi-byte NOP
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 1A /dm */ { 0, BX_IA_NOP }, // multi-byte NOP
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1B /dr */ { 0, BX_IA_NOP }, // multi-byte NOP
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 1B /dm */ { 0, BX_IA_NOP }, // multi-byte NOP
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1C /dr */ { 0, BX_IA_NOP }, // multi-byte NOP
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 1C /dm */ { 0, BX_IA_NOP }, // multi-byte NOP
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1D /dr */ { 0, BX_IA_NOP }, // multi-byte NOP
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 1D /dm */ { 0, BX_IA_NOP }, // multi-byte NOP
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1E /dr */ { 0, BX_IA_NOP }, // multi-byte NOP
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 1E /dm */ { 0, BX_IA_NOP }, // multi-byte NOP
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1F /dr */ { 0, BX_IA_NOP }, // multi-byte NOP
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 1F /dm */ { 0, BX_IA_NOP }, // multi-byte NOP
- prefetch hint should be NOP when use in register mode - #GP when trying to set reserved bits of CR4_HI in 64-bit mode - #GP when trying to set reserved bits of EFER MSR - clear upper part of RSI/RDI when executing rep instructions with 32-bit asize even if no repeat iterations were executed (because of RCX=0 for example) - write SYSENTER_EIP_MSR and SYSENTER_ESP_MSR as 64-bit when x86_64 supported - set MSR_FMASK reset value - MSR_FMASK should be 32-bit only - check for fetch permissions when doing ITLB lookup - #GP when trying to write non-canonical address to MSR_CSTAR or MSR_LSTAR - correct repeat instructions timing - mark TSS busy in TR after it is loaded
2008-04-16 20:44:06 +04:00
#else
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 18 /dr */ { 0, BX_IA_ERROR },
- prefetch hint should be NOP when use in register mode - #GP when trying to set reserved bits of CR4_HI in 64-bit mode - #GP when trying to set reserved bits of EFER MSR - clear upper part of RSI/RDI when executing rep instructions with 32-bit asize even if no repeat iterations were executed (because of RCX=0 for example) - write SYSENTER_EIP_MSR and SYSENTER_ESP_MSR as 64-bit when x86_64 supported - set MSR_FMASK reset value - MSR_FMASK should be 32-bit only - check for fetch permissions when doing ITLB lookup - #GP when trying to write non-canonical address to MSR_CSTAR or MSR_LSTAR - correct repeat instructions timing - mark TSS busy in TR after it is loaded
2008-04-16 20:44:06 +04:00
/* 0F 18 /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 19 /dr */ { 0, BX_IA_ERROR },
- prefetch hint should be NOP when use in register mode - #GP when trying to set reserved bits of CR4_HI in 64-bit mode - #GP when trying to set reserved bits of EFER MSR - clear upper part of RSI/RDI when executing rep instructions with 32-bit asize even if no repeat iterations were executed (because of RCX=0 for example) - write SYSENTER_EIP_MSR and SYSENTER_ESP_MSR as 64-bit when x86_64 supported - set MSR_FMASK reset value - MSR_FMASK should be 32-bit only - check for fetch permissions when doing ITLB lookup - #GP when trying to write non-canonical address to MSR_CSTAR or MSR_LSTAR - correct repeat instructions timing - mark TSS busy in TR after it is loaded
2008-04-16 20:44:06 +04:00
/* 0F 19 /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1A /dr */ { 0, BX_IA_ERROR },
- prefetch hint should be NOP when use in register mode - #GP when trying to set reserved bits of CR4_HI in 64-bit mode - #GP when trying to set reserved bits of EFER MSR - clear upper part of RSI/RDI when executing rep instructions with 32-bit asize even if no repeat iterations were executed (because of RCX=0 for example) - write SYSENTER_EIP_MSR and SYSENTER_ESP_MSR as 64-bit when x86_64 supported - set MSR_FMASK reset value - MSR_FMASK should be 32-bit only - check for fetch permissions when doing ITLB lookup - #GP when trying to write non-canonical address to MSR_CSTAR or MSR_LSTAR - correct repeat instructions timing - mark TSS busy in TR after it is loaded
2008-04-16 20:44:06 +04:00
/* 0F 1A /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1B /dr */ { 0, BX_IA_ERROR },
- prefetch hint should be NOP when use in register mode - #GP when trying to set reserved bits of CR4_HI in 64-bit mode - #GP when trying to set reserved bits of EFER MSR - clear upper part of RSI/RDI when executing rep instructions with 32-bit asize even if no repeat iterations were executed (because of RCX=0 for example) - write SYSENTER_EIP_MSR and SYSENTER_ESP_MSR as 64-bit when x86_64 supported - set MSR_FMASK reset value - MSR_FMASK should be 32-bit only - check for fetch permissions when doing ITLB lookup - #GP when trying to write non-canonical address to MSR_CSTAR or MSR_LSTAR - correct repeat instructions timing - mark TSS busy in TR after it is loaded
2008-04-16 20:44:06 +04:00
/* 0F 1B /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1C /dr */ { 0, BX_IA_ERROR },
- prefetch hint should be NOP when use in register mode - #GP when trying to set reserved bits of CR4_HI in 64-bit mode - #GP when trying to set reserved bits of EFER MSR - clear upper part of RSI/RDI when executing rep instructions with 32-bit asize even if no repeat iterations were executed (because of RCX=0 for example) - write SYSENTER_EIP_MSR and SYSENTER_ESP_MSR as 64-bit when x86_64 supported - set MSR_FMASK reset value - MSR_FMASK should be 32-bit only - check for fetch permissions when doing ITLB lookup - #GP when trying to write non-canonical address to MSR_CSTAR or MSR_LSTAR - correct repeat instructions timing - mark TSS busy in TR after it is loaded
2008-04-16 20:44:06 +04:00
/* 0F 1C /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1D /dr */ { 0, BX_IA_ERROR },
- prefetch hint should be NOP when use in register mode - #GP when trying to set reserved bits of CR4_HI in 64-bit mode - #GP when trying to set reserved bits of EFER MSR - clear upper part of RSI/RDI when executing rep instructions with 32-bit asize even if no repeat iterations were executed (because of RCX=0 for example) - write SYSENTER_EIP_MSR and SYSENTER_ESP_MSR as 64-bit when x86_64 supported - set MSR_FMASK reset value - MSR_FMASK should be 32-bit only - check for fetch permissions when doing ITLB lookup - #GP when trying to write non-canonical address to MSR_CSTAR or MSR_LSTAR - correct repeat instructions timing - mark TSS busy in TR after it is loaded
2008-04-16 20:44:06 +04:00
/* 0F 1D /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1E /dr */ { 0, BX_IA_ERROR },
- prefetch hint should be NOP when use in register mode - #GP when trying to set reserved bits of CR4_HI in 64-bit mode - #GP when trying to set reserved bits of EFER MSR - clear upper part of RSI/RDI when executing rep instructions with 32-bit asize even if no repeat iterations were executed (because of RCX=0 for example) - write SYSENTER_EIP_MSR and SYSENTER_ESP_MSR as 64-bit when x86_64 supported - set MSR_FMASK reset value - MSR_FMASK should be 32-bit only - check for fetch permissions when doing ITLB lookup - #GP when trying to write non-canonical address to MSR_CSTAR or MSR_LSTAR - correct repeat instructions timing - mark TSS busy in TR after it is loaded
2008-04-16 20:44:06 +04:00
/* 0F 1E /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 1F /dr */ { 0, BX_IA_ERROR },
- prefetch hint should be NOP when use in register mode - #GP when trying to set reserved bits of CR4_HI in 64-bit mode - #GP when trying to set reserved bits of EFER MSR - clear upper part of RSI/RDI when executing rep instructions with 32-bit asize even if no repeat iterations were executed (because of RCX=0 for example) - write SYSENTER_EIP_MSR and SYSENTER_ESP_MSR as 64-bit when x86_64 supported - set MSR_FMASK reset value - MSR_FMASK should be 32-bit only - check for fetch permissions when doing ITLB lookup - #GP when trying to write non-canonical address to MSR_CSTAR or MSR_LSTAR - correct repeat instructions timing - mark TSS busy in TR after it is loaded
2008-04-16 20:44:06 +04:00
/* 0F 1F /dm */ { 0, BX_IA_ERROR },
split11b on opcode tables level - split almost eevery splittable instruction will be continued
2007-11-17 15:44:10 +03:00
#endif
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 20 /dr */ { 0, BX_IA_MOV_RdCd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 20 /dm */ { 0, BX_IA_MOV_RdCd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 21 /dr */ { 0, BX_IA_MOV_RdDd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 21 /dm */ { 0, BX_IA_MOV_RdDd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 22 /dr */ { BxTraceEnd, BX_IA_MOV_CdRd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 22 /dm */ { BxTraceEnd, BX_IA_MOV_CdRd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 23 /dr */ { BxTraceEnd, BX_IA_MOV_DdRd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 23 /dm */ { BxTraceEnd, BX_IA_MOV_DdRd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 24 /dr */ { BxTraceEnd, BX_IA_MOV_RdTd }, // not implemented
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 24 /dm */ { BxTraceEnd, BX_IA_MOV_RdTd }, // not implemented
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 25 /dr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 25 /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 26 /dr */ { BxTraceEnd, BX_IA_MOV_TdRd }, // not implemented
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 26 /dm */ { BxTraceEnd, BX_IA_MOV_TdRd }, // not implemented
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 27 /dr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 27 /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 28 /dr */ { BxPrefixSSE, BX_IA_MOVAPS_VpsWpsR, BxOpcodeGroupSSE_0f28R },
split SSE move instructions
2009-10-24 15:17:51 +04:00
/* 0F 28 /dm */ { BxPrefixSSE, BX_IA_MOVAPS_VpsWpsM, BxOpcodeGroupSSE_0f28M },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 29 /dr */ { BxPrefixSSE, BX_IA_MOVAPS_WpsVpsR, BxOpcodeGroupSSE_0f29R },
split SSE move instructions
2009-10-24 15:17:51 +04:00
/* 0F 29 /dm */ { BxPrefixSSE, BX_IA_MOVAPS_WpsVpsM, BxOpcodeGroupSSE_0f29M },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 2A /dr */ { BxPrefixSSE, BX_IA_CVTPI2PS_VpsQqR, BxOpcodeGroupSSE_0f2aR },
split opcodes by ModC0
2009-08-22 15:47:42 +04:00
/* 0F 2A /dm */ { BxPrefixSSE, BX_IA_CVTPI2PS_VpsQqM, BxOpcodeGroupSSE_0f2aM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 2B /dr */ { 0, BX_IA_ERROR }, // MOVNTPS/PD/SS/SD
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 2B /dm */ { BxPrefixSSE, BX_IA_MOVNTPS_MpsVps, BxOpcodeGroupSSE_0f2bM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 2C /dr */ { BxPrefixSSE, BX_IA_CVTTPS2PI_PqWps, BxOpcodeGroupSSE_0f2c },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 2C /dm */ { BxPrefixSSE, BX_IA_CVTTPS2PI_PqWps, BxOpcodeGroupSSE_0f2c },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 2D /dr */ { BxPrefixSSE, BX_IA_CVTPS2PI_PqWps, BxOpcodeGroupSSE_0f2d },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 2D /dm */ { BxPrefixSSE, BX_IA_CVTPS2PI_PqWps, BxOpcodeGroupSSE_0f2d },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 2E /dr */ { BxPrefixSSE, BX_IA_UCOMISS_VssWss, BxOpcodeGroupSSE_0f2e },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 2E /dm */ { BxPrefixSSE, BX_IA_UCOMISS_VssWss, BxOpcodeGroupSSE_0f2e },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 2F /dr */ { BxPrefixSSE, BX_IA_COMISS_VpsWps, BxOpcodeGroupSSE_0f2f },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 2F /dm */ { BxPrefixSSE, BX_IA_COMISS_VpsWps, BxOpcodeGroupSSE_0f2f },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 30 /dr */ { 0, BX_IA_WRMSR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 30 /dm */ { 0, BX_IA_WRMSR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 31 /dr */ { 0, BX_IA_RDTSC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 31 /dm */ { 0, BX_IA_RDTSC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 32 /dr */ { 0, BX_IA_RDMSR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 32 /dm */ { 0, BX_IA_RDMSR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 33 /dr */ { 0, BX_IA_RDPMC },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 33 /dm */ { 0, BX_IA_RDPMC },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 34 /dr */ { BxTraceEnd, BX_IA_SYSENTER },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 34 /dm */ { BxTraceEnd, BX_IA_SYSENTER },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 35 /dr */ { BxTraceEnd, BX_IA_SYSEXIT },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 35 /dm */ { BxTraceEnd, BX_IA_SYSEXIT },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 36 /dr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 36 /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 37 /dr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 37 /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 38 /dr */ { Bx3ByteOp, BX_IA_ERROR, BxOpcode3ByteTable0f38 }, // 3-byte escape
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 38 /dm */ { Bx3ByteOp, BX_IA_ERROR, BxOpcode3ByteTable0f38 }, // 3-byte escape
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 39 /dr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 39 /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 3A /dr */ { Bx3ByteOp | BxImmediate_Ib, BX_IA_ERROR, BxOpcode3ByteTable0f3a }, // 3-byte escape
minimize SSE tables, minor speedup in SSE code
2009-02-27 00:57:01 +03:00
/* 0F 3A /dm */ { Bx3ByteOp | BxImmediate_Ib, BX_IA_ERROR, BxOpcode3ByteTable0f3a }, // 3-byte escape
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 3B /dr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 3B /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 3C /dr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 3C /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 3D /dr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 3D /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 3E /dr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 3E /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 3F /dr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 3F /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 40 /dr */ { 0, BX_IA_CMOVO_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 40 /dm */ { 0, BX_IA_CMOVO_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 41 /dr */ { 0, BX_IA_CMOVNO_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 41 /dm */ { 0, BX_IA_CMOVNO_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 42 /dr */ { 0, BX_IA_CMOVB_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 42 /dm */ { 0, BX_IA_CMOVB_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 43 /dr */ { 0, BX_IA_CMOVNB_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 43 /dm */ { 0, BX_IA_CMOVNB_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 44 /dr */ { 0, BX_IA_CMOVZ_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 44 /dm */ { 0, BX_IA_CMOVZ_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 45 /dr */ { 0, BX_IA_CMOVNZ_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 45 /dm */ { 0, BX_IA_CMOVNZ_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 46 /dr */ { 0, BX_IA_CMOVBE_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 46 /dm */ { 0, BX_IA_CMOVBE_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 47 /dr */ { 0, BX_IA_CMOVNBE_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 47 /dm */ { 0, BX_IA_CMOVNBE_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 48 /dr */ { 0, BX_IA_CMOVS_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 48 /dm */ { 0, BX_IA_CMOVS_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 49 /dr */ { 0, BX_IA_CMOVNS_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 49 /dm */ { 0, BX_IA_CMOVNS_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 4A /dr */ { 0, BX_IA_CMOVP_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 4A /dm */ { 0, BX_IA_CMOVP_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 4B /dr */ { 0, BX_IA_CMOVNP_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 4B /dm */ { 0, BX_IA_CMOVNP_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 4C /dr */ { 0, BX_IA_CMOVL_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 4C /dm */ { 0, BX_IA_CMOVL_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 4D /dr */ { 0, BX_IA_CMOVNL_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 4D /dm */ { 0, BX_IA_CMOVNL_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 4E /dr */ { 0, BX_IA_CMOVLE_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 4E /dm */ { 0, BX_IA_CMOVLE_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 4F /dr */ { 0, BX_IA_CMOVNLE_GdEdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 4F /dm */ { 0, BX_IA_CMOVNLE_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 50 /dr */ { BxPrefixSSE, BX_IA_MOVMSKPS_GdVRps, BxOpcodeGroupSSE_0f50 },
bugfix
2010-03-30 19:01:09 +04:00
/* 0F 50 /dm */ { 0, BX_IA_ERROR }, // MOVMSKPS/PD
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 51 /dr */ { BxPrefixSSE, BX_IA_SQRTPS_VpsWps, BxOpcodeGroupSSE_0f51 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 51 /dm */ { BxPrefixSSE, BX_IA_SQRTPS_VpsWps, BxOpcodeGroupSSE_0f51 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 52 /dr */ { BxPrefixSSE, BX_IA_RSQRTPS_VpsWps, BxOpcodeGroupSSE_0f52 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 52 /dm */ { BxPrefixSSE, BX_IA_RSQRTPS_VpsWps, BxOpcodeGroupSSE_0f52 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 53 /dr */ { BxPrefixSSE, BX_IA_RCPPS_VpsWps, BxOpcodeGroupSSE_0f53 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 53 /dm */ { BxPrefixSSE, BX_IA_RCPPS_VpsWps, BxOpcodeGroupSSE_0f53 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 54 /dr */ { BxPrefixSSE, BX_IA_ANDPS_VpsWps, BxOpcodeGroupSSE_0f54 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 54 /dm */ { BxPrefixSSE, BX_IA_ANDPS_VpsWps, BxOpcodeGroupSSE_0f54 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 55 /dr */ { BxPrefixSSE, BX_IA_ANDNPS_VpsWps, BxOpcodeGroupSSE_0f55 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 55 /dm */ { BxPrefixSSE, BX_IA_ANDNPS_VpsWps, BxOpcodeGroupSSE_0f55 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 56 /dr */ { BxPrefixSSE, BX_IA_ORPS_VpsWps, BxOpcodeGroupSSE_0f56 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 56 /dm */ { BxPrefixSSE, BX_IA_ORPS_VpsWps, BxOpcodeGroupSSE_0f56 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 57 /dr */ { BxPrefixSSE, BX_IA_XORPS_VpsWps, BxOpcodeGroupSSE_0f57 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 57 /dm */ { BxPrefixSSE, BX_IA_XORPS_VpsWps, BxOpcodeGroupSSE_0f57 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 58 /dr */ { BxPrefixSSE, BX_IA_ADDPS_VpsWps, BxOpcodeGroupSSE_0f58 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 58 /dm */ { BxPrefixSSE, BX_IA_ADDPS_VpsWps, BxOpcodeGroupSSE_0f58 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 59 /dr */ { BxPrefixSSE, BX_IA_MULPS_VpsWps, BxOpcodeGroupSSE_0f59 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 59 /dm */ { BxPrefixSSE, BX_IA_MULPS_VpsWps, BxOpcodeGroupSSE_0f59 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 5A /dr */ { BxPrefixSSE, BX_IA_CVTPS2PD_VpsWps, BxOpcodeGroupSSE_0f5a },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 5A /dm */ { BxPrefixSSE, BX_IA_CVTPS2PD_VpsWps, BxOpcodeGroupSSE_0f5a },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 5B /dr */ { BxPrefixSSE, BX_IA_CVTDQ2PS_VpsWdq, BxOpcodeGroupSSE_0f5b },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 5B /dm */ { BxPrefixSSE, BX_IA_CVTDQ2PS_VpsWdq, BxOpcodeGroupSSE_0f5b },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 5C /dr */ { BxPrefixSSE, BX_IA_SUBPS_VpsWps, BxOpcodeGroupSSE_0f5c },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 5C /dm */ { BxPrefixSSE, BX_IA_SUBPS_VpsWps, BxOpcodeGroupSSE_0f5c },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 5D /dr */ { BxPrefixSSE, BX_IA_MINPS_VpsWps, BxOpcodeGroupSSE_0f5d },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 5D /dm */ { BxPrefixSSE, BX_IA_MINPS_VpsWps, BxOpcodeGroupSSE_0f5d },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 5E /dr */ { BxPrefixSSE, BX_IA_DIVPS_VpsWps, BxOpcodeGroupSSE_0f5e },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 5E /dm */ { BxPrefixSSE, BX_IA_DIVPS_VpsWps, BxOpcodeGroupSSE_0f5e },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 5F /dr */ { BxPrefixSSE, BX_IA_MAXPS_VpsWps, BxOpcodeGroupSSE_0f5f },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 5F /dm */ { BxPrefixSSE, BX_IA_MAXPS_VpsWps, BxOpcodeGroupSSE_0f5f },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 60 /dr */ { BxPrefixSSE, BX_IA_PUNPCKLBW_PqQd, BxOpcodeGroupSSE_0f60 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 60 /dm */ { BxPrefixSSE, BX_IA_PUNPCKLBW_PqQd, BxOpcodeGroupSSE_0f60 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 61 /dr */ { BxPrefixSSE, BX_IA_PUNPCKLWD_PqQd, BxOpcodeGroupSSE_0f61 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 61 /dm */ { BxPrefixSSE, BX_IA_PUNPCKLWD_PqQd, BxOpcodeGroupSSE_0f61 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 62 /dr */ { BxPrefixSSE, BX_IA_PUNPCKLDQ_PqQd, BxOpcodeGroupSSE_0f62 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 62 /dm */ { BxPrefixSSE, BX_IA_PUNPCKLDQ_PqQd, BxOpcodeGroupSSE_0f62 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 63 /dr */ { BxPrefixSSE, BX_IA_PACKSSWB_PqQq, BxOpcodeGroupSSE_0f63 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 63 /dm */ { BxPrefixSSE, BX_IA_PACKSSWB_PqQq, BxOpcodeGroupSSE_0f63 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 64 /dr */ { BxPrefixSSE, BX_IA_PCMPGTB_PqQq, BxOpcodeGroupSSE_0f64 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 64 /dm */ { BxPrefixSSE, BX_IA_PCMPGTB_PqQq, BxOpcodeGroupSSE_0f64 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 65 /dr */ { BxPrefixSSE, BX_IA_PCMPGTW_PqQq, BxOpcodeGroupSSE_0f65 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 65 /dm */ { BxPrefixSSE, BX_IA_PCMPGTW_PqQq, BxOpcodeGroupSSE_0f65 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 66 /dr */ { BxPrefixSSE, BX_IA_PCMPGTD_PqQq, BxOpcodeGroupSSE_0f66 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 66 /dm */ { BxPrefixSSE, BX_IA_PCMPGTD_PqQq, BxOpcodeGroupSSE_0f66 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 67 /dr */ { BxPrefixSSE, BX_IA_PACKUSWB_PqQq, BxOpcodeGroupSSE_0f67 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 67 /dm */ { BxPrefixSSE, BX_IA_PACKUSWB_PqQq, BxOpcodeGroupSSE_0f67 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 68 /dr */ { BxPrefixSSE, BX_IA_PUNPCKHBW_PqQq, BxOpcodeGroupSSE_0f68 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 68 /dm */ { BxPrefixSSE, BX_IA_PUNPCKHBW_PqQq, BxOpcodeGroupSSE_0f68 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 69 /dr */ { BxPrefixSSE, BX_IA_PUNPCKHWD_PqQq, BxOpcodeGroupSSE_0f69 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 69 /dm */ { BxPrefixSSE, BX_IA_PUNPCKHWD_PqQq, BxOpcodeGroupSSE_0f69 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 6A /dr */ { BxPrefixSSE, BX_IA_PUNPCKHDQ_PqQq, BxOpcodeGroupSSE_0f6a },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 6A /dm */ { BxPrefixSSE, BX_IA_PUNPCKHDQ_PqQq, BxOpcodeGroupSSE_0f6a },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 6B /dr */ { BxPrefixSSE, BX_IA_PACKSSDW_PqQq, BxOpcodeGroupSSE_0f6b },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 6B /dm */ { BxPrefixSSE, BX_IA_PACKSSDW_PqQq, BxOpcodeGroupSSE_0f6b },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 6C /dr */ { BxPrefixSSE66, BX_IA_PUNPCKLQDQ_VdqWdq },
cleanup
2010-02-08 17:54:26 +03:00
/* 0F 6C /dm */ { BxPrefixSSE66, BX_IA_PUNPCKLQDQ_VdqWdq },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 6D /dr */ { BxPrefixSSE66, BX_IA_PUNPCKHQDQ_VdqWdq },
cleanup
2010-02-08 17:54:26 +03:00
/* 0F 6D /dm */ { BxPrefixSSE66, BX_IA_PUNPCKHQDQ_VdqWdq },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 6E /dr */ { BxPrefixSSE, BX_IA_MOVD_PqEdR, BxOpcodeGroupSSE_0f6eR },
split few more opcodes
2010-02-10 20:21:15 +03:00
/* 0F 6E /dm */ { BxPrefixSSE, BX_IA_MOVD_PqEdM, BxOpcodeGroupSSE_0f6eM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 6F /dr */ { BxPrefixSSE, BX_IA_MOVQ_PqQqR, BxOpcodeGroupSSE_0f6fR },
Split MOVQ_PqQq opcode
2009-12-14 14:55:42 +03:00
/* 0F 6F /dm */ { BxPrefixSSE, BX_IA_MOVQ_PqQqM, BxOpcodeGroupSSE_0f6fM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 70 /dr */ { BxPrefixSSE | BxImmediate_Ib, BX_IA_PSHUFW_PqQqIb, BxOpcodeGroupSSE_0f70 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 70 /dm */ { BxPrefixSSE | BxImmediate_Ib, BX_IA_PSHUFW_PqQqIb, BxOpcodeGroupSSE_0f70 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 71 /dr */ { BxGroup12, BX_IA_ERROR, BxOpcodeInfoG12R },
More fetchdecode fixes
2008-05-01 01:32:33 +04:00
/* 0F 71 /dm */ { 0, BX_IA_ERROR }, // SSE Group G12
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 72 /dr */ { BxGroup13, BX_IA_ERROR, BxOpcodeInfoG13R },
More fetchdecode fixes
2008-05-01 01:32:33 +04:00
/* 0F 72 /dm */ { 0, BX_IA_ERROR }, // SSE Group G13
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 73 /dr */ { BxGroup14, BX_IA_ERROR, BxOpcodeInfoG14R },
More fetchdecode fixes
2008-05-01 01:32:33 +04:00
/* 0F 73 /dm */ { 0, BX_IA_ERROR }, // SSE Group G14
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 74 /dr */ { BxPrefixSSE, BX_IA_PCMPEQB_PqQq, BxOpcodeGroupSSE_0f74 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 74 /dm */ { BxPrefixSSE, BX_IA_PCMPEQB_PqQq, BxOpcodeGroupSSE_0f74 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 75 /dr */ { BxPrefixSSE, BX_IA_PCMPEQW_PqQq, BxOpcodeGroupSSE_0f75 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 75 /dm */ { BxPrefixSSE, BX_IA_PCMPEQW_PqQq, BxOpcodeGroupSSE_0f75 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 76 /dr */ { BxPrefixSSE, BX_IA_PCMPEQD_PqQq, BxOpcodeGroupSSE_0f76 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F 76 /dm */ { BxPrefixSSE, BX_IA_PCMPEQD_PqQq, BxOpcodeGroupSSE_0f76 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 77 /dr */ { BxPrefixSSE, BX_IA_EMMS, BxOpcodeGroupSSE_ERR },
Fixed handling of prefixes for EMMS Small FPU optimization
2008-09-13 00:59:31 +04:00
/* 0F 77 /dm */ { BxPrefixSSE, BX_IA_EMMS, BxOpcodeGroupSSE_ERR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 78 /dr */ { BxPrefixSSE, BX_IA_VMREAD_EdGd, BxOpcodeGroupSSE_ERR },
Added Intel VMX emulation to Bochs CPU
2009-01-31 13:43:24 +03:00
/* 0F 78 /dm */ { BxPrefixSSE, BX_IA_VMREAD_EdGd, BxOpcodeGroupSSE_ERR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 79 /dr */ { BxPrefixSSE, BX_IA_VMWRITE_GdEd, BxOpcodeGroupSSE_ERR },
Added Intel VMX emulation to Bochs CPU
2009-01-31 13:43:24 +03:00
/* 0F 79 /dm */ { BxPrefixSSE, BX_IA_VMWRITE_GdEd, BxOpcodeGroupSSE_ERR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 7A /dr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 7A /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 7B /dr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 7B /dm */ { 0, BX_IA_ERROR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 7C /dr */ { BxPrefixSSE, BX_IA_ERROR, BxOpcodeGroupSSE_0f7c },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 7C /dm */ { BxPrefixSSE, BX_IA_ERROR, BxOpcodeGroupSSE_0f7c },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 7D /dr */ { BxPrefixSSE, BX_IA_ERROR, BxOpcodeGroupSSE_0f7d },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 7D /dm */ { BxPrefixSSE, BX_IA_ERROR, BxOpcodeGroupSSE_0f7d },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 7E /dr */ { BxPrefixSSE, BX_IA_MOVD_EdPdR, BxOpcodeGroupSSE_0f7eR },
split few more opcodes
2010-02-10 20:21:15 +03:00
/* 0F 7E /dm */ { BxPrefixSSE, BX_IA_MOVD_EdPdM, BxOpcodeGroupSSE_0f7eM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 7F /dr */ { BxPrefixSSE, BX_IA_MOVQ_QqPqR, BxOpcodeGroupSSE_0f7fR },
split few more opcodes
2010-02-10 20:21:15 +03:00
/* 0F 7F /dm */ { BxPrefixSSE, BX_IA_MOVQ_QqPqM, BxOpcodeGroupSSE_0f7fM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 80 /dr */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JO_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 80 /dm */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JO_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 81 /dr */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JNO_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 81 /dm */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JNO_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 82 /dr */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JB_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 82 /dm */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JB_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 83 /dr */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JNB_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 83 /dm */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JNB_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 84 /dr */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JZ_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 84 /dm */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JZ_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 85 /dr */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JNZ_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 85 /dm */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JNZ_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 86 /dr */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JBE_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 86 /dm */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JBE_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 87 /dr */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JNBE_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 87 /dm */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JNBE_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 88 /dr */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JS_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 88 /dm */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JS_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 89 /dr */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JNS_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 89 /dm */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JNS_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 8A /dr */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JP_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 8A /dm */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JP_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 8B /dr */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JNP_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 8B /dm */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JNP_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 8C /dr */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JL_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 8C /dm */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JL_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 8D /dr */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JNL_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 8D /dm */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JNL_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 8E /dr */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JLE_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 8E /dm */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JLE_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 8F /dr */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JNLE_Jd },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 8F /dm */ { BxImmediate_BrOff32 | BxTraceJCC, BX_IA_JNLE_Jd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 90 /dr */ { 0, BX_IA_SETO_EbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 90 /dm */ { 0, BX_IA_SETO_EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 91 /dr */ { 0, BX_IA_SETNO_EbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 91 /dm */ { 0, BX_IA_SETNO_EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 92 /dr */ { 0, BX_IA_SETB_EbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 92 /dm */ { 0, BX_IA_SETB_EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 93 /dr */ { 0, BX_IA_SETNB_EbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 93 /dm */ { 0, BX_IA_SETNB_EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 94 /dr */ { 0, BX_IA_SETZ_EbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 94 /dm */ { 0, BX_IA_SETZ_EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 95 /dr */ { 0, BX_IA_SETNZ_EbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 95 /dm */ { 0, BX_IA_SETNZ_EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 96 /dr */ { 0, BX_IA_SETBE_EbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 96 /dm */ { 0, BX_IA_SETBE_EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 97 /dr */ { 0, BX_IA_SETNBE_EbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 97 /dm */ { 0, BX_IA_SETNBE_EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 98 /dr */ { 0, BX_IA_SETS_EbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 98 /dm */ { 0, BX_IA_SETS_EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 99 /dr */ { 0, BX_IA_SETNS_EbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 99 /dm */ { 0, BX_IA_SETNS_EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 9A /dr */ { 0, BX_IA_SETP_EbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 9A /dm */ { 0, BX_IA_SETP_EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 9B /dr */ { 0, BX_IA_SETNP_EbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 9B /dm */ { 0, BX_IA_SETNP_EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 9C /dr */ { 0, BX_IA_SETL_EbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 9C /dm */ { 0, BX_IA_SETL_EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 9D /dr */ { 0, BX_IA_SETNL_EbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 9D /dm */ { 0, BX_IA_SETNL_EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 9E /dr */ { 0, BX_IA_SETLE_EbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 9E /dm */ { 0, BX_IA_SETLE_EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F 9F /dr */ { 0, BX_IA_SETNLE_EbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F 9F /dm */ { 0, BX_IA_SETNLE_EbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A0 /dr */ { 0, BX_IA_PUSH32_FS },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F A0 /dm */ { 0, BX_IA_PUSH32_FS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A1 /dr */ { 0, BX_IA_POP32_FS },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F A1 /dm */ { 0, BX_IA_POP32_FS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A2 /dr */ { 0, BX_IA_CPUID },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F A2 /dm */ { 0, BX_IA_CPUID },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A3 /dr */ { 0, BX_IA_BT_EdGdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F A3 /dm */ { 0, BX_IA_BT_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A4 /dr */ { BxImmediate_Ib, BX_IA_SHLD_EdGdR },
Split SHRD/SHLD instructions
2008-04-05 23:08:01 +04:00
/* 0F A4 /dm */ { BxImmediate_Ib, BX_IA_SHLD_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A5 /dr */ { 0, BX_IA_SHLD_EdGdR },
Split SHRD/SHLD instructions
2008-04-05 23:08:01 +04:00
/* 0F A5 /dm */ { 0, BX_IA_SHLD_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A6 /dr */ { BxTraceEnd, BX_IA_CMPXCHG_XBTS }, // not implemented
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F A6 /dm */ { BxTraceEnd, BX_IA_CMPXCHG_XBTS }, // not implemented
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A7 /dr */ { BxTraceEnd, BX_IA_CMPXCHG_IBTS }, // not implemented
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F A7 /dm */ { BxTraceEnd, BX_IA_CMPXCHG_IBTS }, // not implemented
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A8 /dr */ { 0, BX_IA_PUSH32_GS },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F A8 /dm */ { 0, BX_IA_PUSH32_GS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F A9 /dr */ { 0, BX_IA_POP32_GS },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F A9 /dm */ { 0, BX_IA_POP32_GS },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F AA /dr */ { BxTraceEnd, BX_IA_RSM },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F AA /dm */ { BxTraceEnd, BX_IA_RSM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F AB /dr */ { 0, BX_IA_BTS_EdGdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F AB /dm */ { BxLockable, BX_IA_BTS_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F AC /dr */ { BxImmediate_Ib, BX_IA_SHRD_EdGdR },
Split SHRD/SHLD instructions
2008-04-05 23:08:01 +04:00
/* 0F AC /dm */ { BxImmediate_Ib, BX_IA_SHRD_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F AD /dr */ { 0, BX_IA_SHRD_EdGdR },
Split SHRD/SHLD instructions
2008-04-05 23:08:01 +04:00
/* 0F AD /dm */ { 0, BX_IA_SHRD_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F AE /dr */ { BxGroup15, BX_IA_ERROR, BxOpcodeInfoG15R },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F AE /dm */ { BxGroup15, BX_IA_ERROR, BxOpcodeInfoG15M },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F AF /dr */ { 0, BX_IA_IMUL_GdEdR },
Split more opcodes using new LOAD technique
2008-08-11 01:16:12 +04:00
/* 0F AF /dm */ { 0, BX_IA_IMUL_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B0 /dr */ { 0, BX_IA_CMPXCHG_EbGbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F B0 /dm */ { BxLockable, BX_IA_CMPXCHG_EbGbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B1 /dr */ { 0, BX_IA_CMPXCHG_EdGdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F B1 /dm */ { BxLockable, BX_IA_CMPXCHG_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B2 /dr */ { 0, BX_IA_ERROR }, // LSS
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F B2 /dm */ { 0, BX_IA_LSS_GdMp },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B3 /dr */ { 0, BX_IA_BTR_EdGdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F B3 /dm */ { BxLockable, BX_IA_BTR_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B4 /dr */ { 0, BX_IA_ERROR }, // LFS
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F B4 /dm */ { 0, BX_IA_LFS_GdMp },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B5 /dr */ { 0, BX_IA_ERROR }, // LGS
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F B5 /dm */ { 0, BX_IA_LGS_GdMp },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B6 /dr */ { 0, BX_IA_MOVZX_GdEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F B6 /dm */ { 0, BX_IA_MOVZX_GdEbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B7 /dr */ { 0, BX_IA_MOVZX_GdEwR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F B7 /dm */ { 0, BX_IA_MOVZX_GdEwM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B8 /dr */ { BxPrefixSSEF3, BX_IA_POPCNT_GdEdR },
cleanup
2010-02-08 17:54:26 +03:00
/* 0F B8 /dm */ { BxPrefixSSEF3, BX_IA_POPCNT_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F B9 /dr */ { BxTraceEnd, BX_IA_UD2B },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F B9 /dm */ { BxTraceEnd, BX_IA_UD2B },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F BA /dr */ { BxGroup8, BX_IA_ERROR, BxOpcodeInfoG8EdIbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F BA /dm */ { BxGroup8, BX_IA_ERROR, BxOpcodeInfoG8EdIbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F BB /dr */ { 0, BX_IA_BTC_EdGdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F BB /dm */ { BxLockable, BX_IA_BTC_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F BC /dr */ { 0, BX_IA_BSF_GdEdR },
More use of LOAD_Ex method
2008-08-10 23:34:28 +04:00
/* 0F BC /dm */ { 0, BX_IA_BSF_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F BD /dr */ { 0, BX_IA_BSR_GdEdR },
More use of LOAD_Ex method
2008-08-10 23:34:28 +04:00
/* 0F BD /dm */ { 0, BX_IA_BSR_GdEdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F BE /dr */ { 0, BX_IA_MOVSX_GdEbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F BE /dm */ { 0, BX_IA_MOVSX_GdEbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F BF /dr */ { 0, BX_IA_MOVSX_GdEwR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F BF /dm */ { 0, BX_IA_MOVSX_GdEwM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C0 /dr */ { 0, BX_IA_XADD_EbGbR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F C0 /dm */ { BxLockable, BX_IA_XADD_EbGbM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C1 /dr */ { 0, BX_IA_XADD_EdGdR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F C1 /dm */ { BxLockable, BX_IA_XADD_EdGdM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C2 /dr */ { BxPrefixSSE | BxImmediate_Ib, BX_IA_CMPPS_VpsWpsIb, BxOpcodeGroupSSE_0fc2 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F C2 /dm */ { BxPrefixSSE | BxImmediate_Ib, BX_IA_CMPPS_VpsWpsIb, BxOpcodeGroupSSE_0fc2 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C3 /dr */ { 0, BX_IA_ERROR }, // MOVNTI SSE group
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F C3 /dm */ { BxPrefixSSE, BX_IA_MOVNTI_MdGd, BxOpcodeGroupSSE_ERR },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C4 /dr */ { BxPrefixSSE | BxImmediate_Ib, BX_IA_PINSRW_PqEwIb, BxOpcodeGroupSSE_0fc4 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F C4 /dm */ { BxPrefixSSE | BxImmediate_Ib, BX_IA_PINSRW_PqEwIb, BxOpcodeGroupSSE_0fc4 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C5 /dr */ { BxPrefixSSE | BxImmediate_Ib, BX_IA_PEXTRW_GdPqIb, BxOpcodeGroupSSE_0fc5R },
some fetchdecode fixes
2008-05-01 01:07:12 +04:00
/* 0F C5 /dm */ { 0, BX_IA_ERROR }, // PEXTRW SSE group
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C6 /dr */ { BxPrefixSSE | BxImmediate_Ib, BX_IA_SHUFPS_VpsWpsIb, BxOpcodeGroupSSE_0fc6 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F C6 /dm */ { BxPrefixSSE | BxImmediate_Ib, BX_IA_SHUFPS_VpsWpsIb, BxOpcodeGroupSSE_0fc6 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C7 /dr */ { 0, BX_IA_ERROR }, // CMPXCHG8B G9 group
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F C7 /dm */ { BxGroup9, BX_IA_ERROR, BxOpcodeInfoG9M },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C8 /dr */ { 0, BX_IA_BSWAP_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F C8 /dm */ { 0, BX_IA_BSWAP_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F C9 /dr */ { 0, BX_IA_BSWAP_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F C9 /dm */ { 0, BX_IA_BSWAP_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F CA /dr */ { 0, BX_IA_BSWAP_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F CA /dm */ { 0, BX_IA_BSWAP_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F CB /dr */ { 0, BX_IA_BSWAP_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F CB /dm */ { 0, BX_IA_BSWAP_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F CC /dr */ { 0, BX_IA_BSWAP_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F CC /dm */ { 0, BX_IA_BSWAP_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F CD /dr */ { 0, BX_IA_BSWAP_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F CD /dm */ { 0, BX_IA_BSWAP_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F CE /dr */ { 0, BX_IA_BSWAP_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F CE /dm */ { 0, BX_IA_BSWAP_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F CF /dr */ { 0, BX_IA_BSWAP_ERX },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F CF /dm */ { 0, BX_IA_BSWAP_ERX },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D0 /dr */ { BxPrefixSSE, BX_IA_ERROR, BxOpcodeGroupSSE_0fd0 },
Fixed merge error
2008-08-30 00:43:05 +04:00
/* 0F D0 /dm */ { BxPrefixSSE, BX_IA_ERROR, BxOpcodeGroupSSE_0fd0 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D1 /dr */ { BxPrefixSSE, BX_IA_PSRLW_PqQq, BxOpcodeGroupSSE_0fd1 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F D1 /dm */ { BxPrefixSSE, BX_IA_PSRLW_PqQq, BxOpcodeGroupSSE_0fd1 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D2 /dr */ { BxPrefixSSE, BX_IA_PSRLD_PqQq, BxOpcodeGroupSSE_0fd2 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F D2 /dm */ { BxPrefixSSE, BX_IA_PSRLD_PqQq, BxOpcodeGroupSSE_0fd2 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D3 /dr */ { BxPrefixSSE, BX_IA_PSRLQ_PqQq, BxOpcodeGroupSSE_0fd3 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F D3 /dm */ { BxPrefixSSE, BX_IA_PSRLQ_PqQq, BxOpcodeGroupSSE_0fd3 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D4 /dr */ { BxPrefixSSE, BX_IA_PADDQ_PqQq, BxOpcodeGroupSSE_0fd4 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F D4 /dm */ { BxPrefixSSE, BX_IA_PADDQ_PqQq, BxOpcodeGroupSSE_0fd4 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D5 /dr */ { BxPrefixSSE, BX_IA_PMULLW_PqQq, BxOpcodeGroupSSE_0fd5 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F D5 /dm */ { BxPrefixSSE, BX_IA_PMULLW_PqQq, BxOpcodeGroupSSE_0fd5 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D6 /dr */ { BxPrefixSSE, BX_IA_ERROR, BxOpcodeGroupSSE_0fd6R },
bugfix
2010-03-30 19:01:09 +04:00
/* 0F D6 /dm */ { BxPrefixSSE66, BX_IA_MOVQ_WqVqM },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D7 /dr */ { BxPrefixSSE, BX_IA_PMOVMSKB_GdPRq, BxOpcodeGroupSSE_0fd7 },
bugfix
2010-03-30 19:01:09 +04:00
/* 0F D7 /dm */ { 0, BX_IA_ERROR }, // PMOVMSKB is reg/reg form only
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D8 /dr */ { BxPrefixSSE, BX_IA_PSUBUSB_PqQq, BxOpcodeGroupSSE_0fd8 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F D8 /dm */ { BxPrefixSSE, BX_IA_PSUBUSB_PqQq, BxOpcodeGroupSSE_0fd8 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F D9 /dr */ { BxPrefixSSE, BX_IA_PSUBUSW_PqQq, BxOpcodeGroupSSE_0fd9 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F D9 /dm */ { BxPrefixSSE, BX_IA_PSUBUSW_PqQq, BxOpcodeGroupSSE_0fd9 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F DA /dr */ { BxPrefixSSE, BX_IA_PMINUB_PqQq, BxOpcodeGroupSSE_0fda },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F DA /dm */ { BxPrefixSSE, BX_IA_PMINUB_PqQq, BxOpcodeGroupSSE_0fda },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F DB /dr */ { BxPrefixSSE, BX_IA_PAND_PqQq, BxOpcodeGroupSSE_0fdb },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F DB /dm */ { BxPrefixSSE, BX_IA_PAND_PqQq, BxOpcodeGroupSSE_0fdb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F DC /dr */ { BxPrefixSSE, BX_IA_PADDUSB_PqQq, BxOpcodeGroupSSE_0fdc },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F DC /dm */ { BxPrefixSSE, BX_IA_PADDUSB_PqQq, BxOpcodeGroupSSE_0fdc },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F DD /dr */ { BxPrefixSSE, BX_IA_PADDUSW_PqQq, BxOpcodeGroupSSE_0fdd },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F DD /dm */ { BxPrefixSSE, BX_IA_PADDUSW_PqQq, BxOpcodeGroupSSE_0fdd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F DE /dr */ { BxPrefixSSE, BX_IA_PMAXUB_PqQq, BxOpcodeGroupSSE_0fde },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F DE /dm */ { BxPrefixSSE, BX_IA_PMAXUB_PqQq, BxOpcodeGroupSSE_0fde },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F DF /dr */ { BxPrefixSSE, BX_IA_PANDN_PqQq, BxOpcodeGroupSSE_0fdf },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F DF /dm */ { BxPrefixSSE, BX_IA_PANDN_PqQq, BxOpcodeGroupSSE_0fdf },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E0 /dr */ { BxPrefixSSE, BX_IA_PAVGB_PqQq, BxOpcodeGroupSSE_0fe0 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F E0 /dm */ { BxPrefixSSE, BX_IA_PAVGB_PqQq, BxOpcodeGroupSSE_0fe0 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E1 /dr */ { BxPrefixSSE, BX_IA_PSRAW_PqQq, BxOpcodeGroupSSE_0fe1 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F E1 /dm */ { BxPrefixSSE, BX_IA_PSRAW_PqQq, BxOpcodeGroupSSE_0fe1 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E2 /dr */ { BxPrefixSSE, BX_IA_PSRAD_PqQq, BxOpcodeGroupSSE_0fe2 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F E2 /dm */ { BxPrefixSSE, BX_IA_PSRAD_PqQq, BxOpcodeGroupSSE_0fe2 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E3 /dr */ { BxPrefixSSE, BX_IA_PAVGW_PqQq, BxOpcodeGroupSSE_0fe3 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F E3 /dm */ { BxPrefixSSE, BX_IA_PAVGW_PqQq, BxOpcodeGroupSSE_0fe3 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E4 /dr */ { BxPrefixSSE, BX_IA_PMULHUW_PqQq, BxOpcodeGroupSSE_0fe4 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F E4 /dm */ { BxPrefixSSE, BX_IA_PMULHUW_PqQq, BxOpcodeGroupSSE_0fe4 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E5 /dr */ { BxPrefixSSE, BX_IA_PMULHW_PqQq, BxOpcodeGroupSSE_0fe5 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F E5 /dm */ { BxPrefixSSE, BX_IA_PMULHW_PqQq, BxOpcodeGroupSSE_0fe5 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E6 /dr */ { BxPrefixSSE, BX_IA_ERROR, BxOpcodeGroupSSE_0fe6 },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F E6 /dm */ { BxPrefixSSE, BX_IA_ERROR, BxOpcodeGroupSSE_0fe6 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E7 /dr */ { 0, BX_IA_ERROR }, // MOVNTQ/MOVNTDQ SSE group
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F E7 /dm */ { BxPrefixSSE, BX_IA_MOVNTQ_MqPq, BxOpcodeGroupSSE_0fe7M },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E8 /dr */ { BxPrefixSSE, BX_IA_PSUBSB_PqQq, BxOpcodeGroupSSE_0fe8 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F E8 /dm */ { BxPrefixSSE, BX_IA_PSUBSB_PqQq, BxOpcodeGroupSSE_0fe8 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F E9 /dr */ { BxPrefixSSE, BX_IA_PSUBSW_PqQq, BxOpcodeGroupSSE_0fe9 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F E9 /dm */ { BxPrefixSSE, BX_IA_PSUBSW_PqQq, BxOpcodeGroupSSE_0fe9 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F EA /dr */ { BxPrefixSSE, BX_IA_PMINSW_PqQq, BxOpcodeGroupSSE_0fea },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F EA /dm */ { BxPrefixSSE, BX_IA_PMINSW_PqQq, BxOpcodeGroupSSE_0fea },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F EB /dr */ { BxPrefixSSE, BX_IA_POR_PqQq, BxOpcodeGroupSSE_0feb },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F EB /dm */ { BxPrefixSSE, BX_IA_POR_PqQq, BxOpcodeGroupSSE_0feb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F EC /dr */ { BxPrefixSSE, BX_IA_PADDSB_PqQq, BxOpcodeGroupSSE_0fec },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F EC /dm */ { BxPrefixSSE, BX_IA_PADDSB_PqQq, BxOpcodeGroupSSE_0fec },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F ED /dr */ { BxPrefixSSE, BX_IA_PADDSW_PqQq, BxOpcodeGroupSSE_0fed },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F ED /dm */ { BxPrefixSSE, BX_IA_PADDSW_PqQq, BxOpcodeGroupSSE_0fed },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F EE /dr */ { BxPrefixSSE, BX_IA_PMAXSW_PqQq, BxOpcodeGroupSSE_0fee },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F EE /dm */ { BxPrefixSSE, BX_IA_PMAXSW_PqQq, BxOpcodeGroupSSE_0fee },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F EF /dr */ { BxPrefixSSE, BX_IA_PXOR_PqQq, BxOpcodeGroupSSE_0fef },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F EF /dm */ { BxPrefixSSE, BX_IA_PXOR_PqQq, BxOpcodeGroupSSE_0fef },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F0 /dr */ { 0, BX_IA_ERROR }, // LDDQU SSE group
fixes and cleanups in disasm and decoder
2010-02-09 22:44:25 +03:00
/* 0F F0 /wm */ { BxPrefixSSEF2, BX_IA_LDDQU_VdqMdq },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F1 /dr */ { BxPrefixSSE, BX_IA_PSLLW_PqQq, BxOpcodeGroupSSE_0ff1 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F F1 /dm */ { BxPrefixSSE, BX_IA_PSLLW_PqQq, BxOpcodeGroupSSE_0ff1 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F2 /dr */ { BxPrefixSSE, BX_IA_PSLLD_PqQq, BxOpcodeGroupSSE_0ff2 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F F2 /dm */ { BxPrefixSSE, BX_IA_PSLLD_PqQq, BxOpcodeGroupSSE_0ff2 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F3 /dr */ { BxPrefixSSE, BX_IA_PSLLQ_PqQq, BxOpcodeGroupSSE_0ff3 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F F3 /dm */ { BxPrefixSSE, BX_IA_PSLLQ_PqQq, BxOpcodeGroupSSE_0ff3 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F4 /dr */ { BxPrefixSSE, BX_IA_PMULUDQ_PqQq, BxOpcodeGroupSSE_0ff4 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F F4 /dm */ { BxPrefixSSE, BX_IA_PMULUDQ_PqQq, BxOpcodeGroupSSE_0ff4 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F5 /dr */ { BxPrefixSSE, BX_IA_PMADDWD_PqQq, BxOpcodeGroupSSE_0ff5 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F F5 /dm */ { BxPrefixSSE, BX_IA_PMADDWD_PqQq, BxOpcodeGroupSSE_0ff5 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F6 /dr */ { BxPrefixSSE, BX_IA_PSADBW_PqQq, BxOpcodeGroupSSE_0ff6 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F F6 /dm */ { BxPrefixSSE, BX_IA_PSADBW_PqQq, BxOpcodeGroupSSE_0ff6 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F7 /dr */ { BxPrefixSSE, BX_IA_MASKMOVQ_PqPRq, BxOpcodeGroupSSE_0ff7R },
Complete MASKMOVDQU decoding fix
2008-04-16 09:56:55 +04:00
/* 0F F7 /dm */ { 0, BX_IA_ERROR }, // MASKMOV SSE group
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F8 /dr */ { BxPrefixSSE, BX_IA_PSUBB_PqQq, BxOpcodeGroupSSE_0ff8 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F F8 /dm */ { BxPrefixSSE, BX_IA_PSUBB_PqQq, BxOpcodeGroupSSE_0ff8 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F F9 /dr */ { BxPrefixSSE, BX_IA_PSUBW_PqQq, BxOpcodeGroupSSE_0ff9 },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F F9 /dm */ { BxPrefixSSE, BX_IA_PSUBW_PqQq, BxOpcodeGroupSSE_0ff9 },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F FA /dr */ { BxPrefixSSE, BX_IA_PSUBD_PqQq, BxOpcodeGroupSSE_0ffa },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F FA /dm */ { BxPrefixSSE, BX_IA_PSUBD_PqQq, BxOpcodeGroupSSE_0ffa },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F FB /dr */ { BxPrefixSSE, BX_IA_PSUBQ_PqQq, BxOpcodeGroupSSE_0ffb },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F FB /dm */ { BxPrefixSSE, BX_IA_PSUBQ_PqQq, BxOpcodeGroupSSE_0ffb },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F FC /dr */ { BxPrefixSSE, BX_IA_PADDB_PqQq, BxOpcodeGroupSSE_0ffc },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F FC /dm */ { BxPrefixSSE, BX_IA_PADDB_PqQq, BxOpcodeGroupSSE_0ffc },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F FD /dr */ { BxPrefixSSE, BX_IA_PADDW_PqQq, BxOpcodeGroupSSE_0ffd },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F FD /dm */ { BxPrefixSSE, BX_IA_PADDW_PqQq, BxOpcodeGroupSSE_0ffd },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F FE /dr */ { BxPrefixSSE, BX_IA_PADDD_PqQq, BxOpcodeGroupSSE_0ffe },
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* 0F FE /dm */ { BxPrefixSSE, BX_IA_PADDD_PqQq, BxOpcodeGroupSSE_0ffe },
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
/* 0F FF /dr */ { 0, BX_IA_ERROR },
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
/* 0F FF /dm */ { 0, BX_IA_ERROR }
Fixed CALL/JMP far through call gate 64 Decode SWAPGS and RDTSCP instructions Indent changes in fetchdecode
2006-03-22 23:47:11 +03:00
};
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
small optimization
2009-12-21 16:38:06 +03:00
int BX_CPP_AttrRegparmN(3)
Added const to fetchPtr in cpu functions
2008-03-03 18:16:46 +03:00
BX_CPU_C::fetchDecode32(const Bit8u *iptr, bxInstruction_c *i, unsigned remainingInPage)
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
{
// remain must be at least 1
Move MaxFetch calculation into fetchdecode - simplify the logic
2007-12-22 20:17:40 +03:00
unsigned remain = (remainingInPage < 15) ? remainingInPage : 15;
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
Merged patch-unallowed-lock-cases patch. According to the Intel manuals: The LOCK prefix can be prepended only to the following instructions and only to those forms of the instructions where the destination operand is a memory operand: ADD, ADC, AND, BTC, BTR, BTS, CMPXCHG, CMPXCH8B, DEC, INC, NEG, NOT, OR, SBB, SUB, XOR, XADD, and XCHG. If the LOCK prefix is used with one of these instructions and the source operand is a memory operand, an undefined opcode exception (#UD) will be generated. An undefined opcode exception will also be generated if the LOCK prefix is used with any instruction not in the above list. Checking of the LOCK prefix done in fetchDecode state and not overloads Bochs's execution.
2003-04-05 16:16:53 +04:00
bx_bool is_32, lock=0;
Inline pop/push functions Store only single byte of opcode in b1() - speedup shift instructions Code cleanups
2008-04-05 21:51:55 +04:00
unsigned b1, b2, ilen=0, attr, os_32, ia_opcode = 0;
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
unsigned imm_mode, offset, index = 0;
Added decoding of new SSE4 instructions (recently published in Intel docs) At least CPUID detects them correctly The code is never tested (still) ! (but should work fine)
2006-04-06 22:30:05 +04:00
unsigned rm = 0, mod=0, nnn=0;
optimize seg_override decoding
2008-09-08 20:15:59 +04:00
unsigned seg = BX_SEG_REG_DS, seg_override = BX_SEG_REG_NULL;
Changes method of resolving opcode/attributes from group table New method more flexible and easy to understanding. Reorganizing fetchdecode code and make it more easy and understandable
2003-08-28 23:25:23 +04:00
Merged BOCHS-SSE branch
2002-10-16 21:37:35 +04:00
#define SSE_PREFIX_NONE 0
#define SSE_PREFIX_66 1
#define SSE_PREFIX_F2 2
Decoding : F2 and F2 prefix could override prefix 66 when determine SSE opcode
2007-10-20 14:56:44 +04:00
#define SSE_PREFIX_F3 3
- Code cleanup, remove x86-64 code from functions which cannot be called from x86-64 - Fix PANIC multiple SSE prefix decoding (fetchdecode and disasm) - More Bit32u -> bx_phy_address convert - Lazy flags optimization
2006-05-12 21:04:19 +04:00
unsigned sse_prefix = SSE_PREFIX_NONE;
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
Split out JZ/JNZ instructions from JCC because they were called so frequently. Coded asm() statements for INC/DEC_ERX() instructions. Cleaned up the iCache a litle including a bug fix. The generation ID was decrementing the whole field including some high meta bits. That could roll over after 1 Billion cycles. I know only decrement if the field is valid, to save the write. I implemented inline functions which can serve the value of the arithmetic flags if they are cached, and redirect to the lazy_flags.cc routines if not. Most of this was just prep work for adding more asm() statements for native eflags processing when on x86.
2002-09-22 05:52:21 +04:00
os_32 = is_32 =
BX_CPU_THIS_PTR sregs[BX_SEG_REG_CS].cache.u.segment.d_b;
Eliminate BxResolve call from the heart of cpu loop and move into instructions that really require this calculation. Yes, it blows the code of EVERY CPU method but it has >15% speedup !
2008-01-10 22:37:56 +03:00
i->ResolveModrm = 0;
Unpack more fields in bxInstruction_c -> this increase bxInstruction size by 4 bytes but I have no way but do it if want to support SSE5 dest override later
2008-02-05 00:28:53 +03:00
i->init(/*os32*/ is_32, /*as32*/ is_32,
/*os64*/ 0, /*as64*/ 0);
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
Simplify and optimize fetchdecode methods. Now fetchdecode is simpler to understand and easier to modify, for example to support 3-byte opcodes (SSE4)
2006-04-06 00:52:40 +04:00
offset = os_32 << 9; // * 512
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
fetch_b1:
b1 = *iptr++;
Simplify and optimize fetchdecode methods. Now fetchdecode is simpler to understand and easier to modify, for example to support 3-byte opcodes (SSE4)
2006-04-06 00:52:40 +04:00
ilen++;
Committed patches/patch-bochs-instrumentation from Stanislav. Some things changed in the ctrl_xfer*.cc, fetchdecode*.cc, and cpu.cc since the original patches, so I did some patch integration by hand. Check the placement of the macros BX_INSTR_FETCH_DECODE_COMPLETED() and BX_INSTR_OPCODE() in cpu.cc to make sure I go them right. Also, I changed the parameters to BX_INSTR_OPCODE() to update them to the new code. I put some comments before each of these to help determine if the placement is right. These macros are only compiled in if you are gathering instrumentation data from bochs, so they shouldn't effect others.
2002-09-28 04:54:05 +04:00
Optimized memory access for 64-bit mode Starting convergence to new lazy flags scheme by Darek Mihocka (www.emulators.com). The new flags code is still being validated and perfected but I try to minimize the diff between 2 versionS
2007-11-20 20:15:33 +03:00
switch (b1) {
Small code optimization
2008-06-09 23:35:59 +04:00
case 0x0f: // 2-byte escape
if (ilen < remain) {
ilen++;
b1 = 0x100 | *iptr++;
break;
}
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
Optimized memory access for 64-bit mode Starting convergence to new lazy flags scheme by Darek Mihocka (www.emulators.com). The new flags code is still being validated and perfected but I try to minimize the diff between 2 versionS
2007-11-20 20:15:33 +03:00
case 0x66: // OpSize
os_32 = !is_32;
offset = os_32 << 9;
if(!sse_prefix) sse_prefix = SSE_PREFIX_66;
More optimization for string instructions
2007-12-17 22:52:01 +03:00
i->setOs32B(os_32);
Optimized memory access for 64-bit mode Starting convergence to new lazy flags scheme by Darek Mihocka (www.emulators.com). The new flags code is still being validated and perfected but I try to minimize the diff between 2 versionS
2007-11-20 20:15:33 +03:00
if (ilen < remain) {
goto fetch_b1;
}
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
Optimized memory access for 64-bit mode Starting convergence to new lazy flags scheme by Darek Mihocka (www.emulators.com). The new flags code is still being validated and perfected but I try to minimize the diff between 2 versionS
2007-11-20 20:15:33 +03:00
case 0x67: // AddrSize
More optimization for string instructions
2007-12-17 22:52:01 +03:00
i->setAs32B(!is_32);
Optimized memory access for 64-bit mode Starting convergence to new lazy flags scheme by Darek Mihocka (www.emulators.com). The new flags code is still being validated and perfected but I try to minimize the diff between 2 versionS
2007-11-20 20:15:33 +03:00
if (ilen < remain) {
goto fetch_b1;
}
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
Optimized memory access for 64-bit mode Starting convergence to new lazy flags scheme by Darek Mihocka (www.emulators.com). The new flags code is still being validated and perfected but I try to minimize the diff between 2 versionS
2007-11-20 20:15:33 +03:00
case 0xf2: // REPNE/REPNZ
case 0xf3: // REP/REPE/REPZ
small optimization
2009-11-06 21:19:01 +03:00
sse_prefix = b1 & 3;
More optimization for string instructions
2007-12-17 22:52:01 +03:00
i->setRepUsed(b1 & 3);
Optimized memory access for 64-bit mode Starting convergence to new lazy flags scheme by Darek Mihocka (www.emulators.com). The new flags code is still being validated and perfected but I try to minimize the diff between 2 versionS
2007-11-20 20:15:33 +03:00
if (ilen < remain) {
goto fetch_b1;
}
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
Optimized memory access for 64-bit mode Starting convergence to new lazy flags scheme by Darek Mihocka (www.emulators.com). The new flags code is still being validated and perfected but I try to minimize the diff between 2 versionS
2007-11-20 20:15:33 +03:00
case 0x26: // ES:
Fixed code duplication during prefix decoding
2008-06-12 00:58:29 +04:00
case 0x2e: // CS:
Optimized memory access for 64-bit mode Starting convergence to new lazy flags scheme by Darek Mihocka (www.emulators.com). The new flags code is still being validated and perfected but I try to minimize the diff between 2 versionS
2007-11-20 20:15:33 +03:00
case 0x36: // SS:
case 0x3e: // DS:
optimize seg_override decoding
2008-09-08 20:15:59 +04:00
seg_override = (b1 >> 3) & 3;
Optimized memory access for 64-bit mode Starting convergence to new lazy flags scheme by Darek Mihocka (www.emulators.com). The new flags code is still being validated and perfected but I try to minimize the diff between 2 versionS
2007-11-20 20:15:33 +03:00
if (ilen < remain) {
goto fetch_b1;
}
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
Optimized memory access for 64-bit mode Starting convergence to new lazy flags scheme by Darek Mihocka (www.emulators.com). The new flags code is still being validated and perfected but I try to minimize the diff between 2 versionS
2007-11-20 20:15:33 +03:00
case 0x64: // FS:
case 0x65: // GS:
optimize seg_override decoding
2008-09-08 20:15:59 +04:00
seg_override = (b1 & 0xf);
Optimized memory access for 64-bit mode Starting convergence to new lazy flags scheme by Darek Mihocka (www.emulators.com). The new flags code is still being validated and perfected but I try to minimize the diff between 2 versionS
2007-11-20 20:15:33 +03:00
if (ilen < remain) {
goto fetch_b1;
}
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
Optimized memory access for 64-bit mode Starting convergence to new lazy flags scheme by Darek Mihocka (www.emulators.com). The new flags code is still being validated and perfected but I try to minimize the diff between 2 versionS
2007-11-20 20:15:33 +03:00
case 0xf0: // LOCK:
lock = 1;
if (ilen < remain) {
goto fetch_b1;
}
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
Optimized memory access for 64-bit mode Starting convergence to new lazy flags scheme by Darek Mihocka (www.emulators.com). The new flags code is still being validated and perfected but I try to minimize the diff between 2 versionS
2007-11-20 20:15:33 +03:00
default:
break;
Simplify and optimize fetchdecode methods. Now fetchdecode is simpler to understand and easier to modify, for example to support 3-byte opcodes (SSE4)
2006-04-06 00:52:40 +04:00
}
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
index = (b1+offset) << 1;
attr = BxOpcodeInfo32[index].Attr;
Simplify and optimize fetchdecode methods. Now fetchdecode is simpler to understand and easier to modify, for example to support 3-byte opcodes (SSE4)
2006-04-06 00:52:40 +04:00
Move 3byte opcode decoding under Modrm condition
2007-11-16 23:49:51 +03:00
if (BxOpcodeHasModrm32[b1]) {
unsigned b3 = 0;
// handle 3-byte escape
Fixed problem introduced by prev checkin + Fix beak to debugger when executing HLT instruction
2008-02-29 08:39:40 +03:00
if ((attr & BxGroupX) == Bx3ByteOp) {
Move 3byte opcode decoding under Modrm condition
2007-11-16 23:49:51 +03:00
if (ilen < remain) {
ilen++;
b3 = *iptr++;
}
else
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
MMX - modify masked write to RMW - faster execution CMPXCHG8B/16B - fixed possible problem. Instruction not allowed to fault after some part of it written to the memory
2007-12-20 02:21:11 +03:00
}
Added decoding of new SSE4 instructions (recently published in Intel docs) At least CPUID detects them correctly The code is never tested (still) ! (but should work fine)
2006-04-06 22:30:05 +04:00
Simplify and optimize fetchdecode methods. Now fetchdecode is simpler to understand and easier to modify, for example to support 3-byte opcodes (SSE4)
2006-04-06 00:52:40 +04:00
// opcode requires modrm byte
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
if (ilen < remain) {
ilen++;
b2 = *iptr++;
In 64-bit mode, the CS, DS, ES, and SS segment overrides are ignored.
2005-02-12 22:25:33 +03:00
}
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
else
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
// Parse mod-nnn-rm and related bytes
Simplify and optimize fetchdecode methods. Now fetchdecode is simpler to understand and easier to modify, for example to support 3-byte opcodes (SSE4)
2006-04-06 00:52:40 +04:00
mod = b2 & 0xc0; // leave unshifted
optimize modrm/sib decoding
2007-11-19 00:38:58 +03:00
nnn = (b2 >> 3) & 0x7;
rm = b2 & 0x7;
Updated accessing of modrm/sib addressing information to use accessors. This lets me work on compressing the size of fetch-decode structure (now called bxInstruction_c). I've reduced it down to about 76 bytes. We should be able to do much better soon. I needed the abstraction of the accessors, so I have a lot of freedom to re-arrange things without making massive future changes. Lost a few percent of performance in these mods, but my main focus was to get the abstraction.
2002-09-18 02:50:53 +04:00
CPU fixes by Kevin Lawton
2005-02-17 00:27:21 +03:00
// MOVs with CRx and DRx always use register ops and ignore the mod field.
Simplify and optimize fetchdecode methods. Now fetchdecode is simpler to understand and easier to modify, for example to support 3-byte opcodes (SSE4)
2006-04-06 00:52:40 +04:00
if ((b1 & ~3) == 0x120)
CPU fixes by Kevin Lawton
2005-02-17 00:27:21 +03:00
mod = 0xc0;
Unpack more fields in bxInstruction_c -> this increase bxInstruction size by 4 bytes but I have no way but do it if want to support SSE5 dest override later
2008-02-05 00:28:53 +03:00
i->setModRM(b2);
Organize bxInstruction fields differently
2008-03-31 21:33:34 +04:00
i->setNnn(nnn);
I merged and succeded to remove some similar execution functions - less code, less chance for branch misprediction
2008-01-25 22:34:30 +03:00
Updated accessing of modrm/sib addressing information to use accessors. This lets me work on compressing the size of fetch-decode structure (now called bxInstruction_c). I've reduced it down to about 76 bytes. We should be able to do much better soon. I needed the abstraction of the accessors, so I have a lot of freedom to re-arrange things without making massive future changes. Lost a few percent of performance in these mods, but my main focus was to get the abstraction.
2002-09-18 02:50:53 +04:00
if (mod == 0xc0) { // mod == 11b
More optimization for string instructions
2007-12-17 22:52:01 +03:00
i->assertModC0();
Completely new way to handle LD+OP cases - allows to significantly reduce number of BX_CPU_C methods
2008-08-10 01:05:07 +04:00
i->setRm(rm);
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
goto modrm_done;
Merged patch-unallowed-lock-cases patch. According to the Intel manuals: The LOCK prefix can be prepended only to the following instructions and only to those forms of the instructions where the destination operand is a memory operand: ADD, ADC, AND, BTC, BTR, BTS, CMPXCHG, CMPXCH8B, DEC, INC, NEG, NOT, OR, SBB, SUB, XOR, XADD, and XCHG. If the LOCK prefix is used with one of these instructions and the source operand is a memory operand, an undefined opcode exception (#UD) will be generated. An undefined opcode exception will also be generated if the LOCK prefix is used with any instruction not in the above list. Checking of the LOCK prefix done in fetchDecode state and not overloads Bochs's execution.
2003-04-05 16:16:53 +04:00
}
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
index++; // /m form
Completely new way to handle LD+OP cases - allows to significantly reduce number of BX_CPU_C methods
2008-08-10 01:05:07 +04:00
i->setRm(BX_TMP_REGISTER);
Add NIL register and simplify more BxResolve work
2008-03-29 12:34:35 +03:00
i->setSibBase(rm); // initialize with rm to use BxResolve32Base
preparations for future optimizations - not necessary speedupo now
2008-04-03 21:56:59 +04:00
i->setSibIndex(BX_NIL_REGISTER);
Add NIL register and simplify more BxResolve work
2008-03-29 12:34:35 +03:00
// initialize displ32 with zero to include cases with no diplacement
i->modRMForm.displ32u = 0;
More optimization for string instructions
2007-12-17 22:52:01 +03:00
if (i->as32L()) {
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
// 32-bit addressing modes; note that mod==11b handled above
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
i->ResolveModrm = &BX_CPU_C::BxResolve32Base;
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
if (rm != 4) { // no s-i-b byte
Updated accessing of modrm/sib addressing information to use accessors. This lets me work on compressing the size of fetch-decode structure (now called bxInstruction_c). I've reduced it down to about 76 bytes. We should be able to do much better soon. I needed the abstraction of the accessors, so I have a lot of freedom to re-arrange things without making massive future changes. Lost a few percent of performance in these mods, but my main focus was to get the abstraction.
2002-09-18 02:50:53 +04:00
if (mod == 0x00) { // mod == 00b
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
if (rm == 5) {
preparations for future optimizations - not necessary speedupo now
2008-04-03 21:56:59 +04:00
i->setSibBase(BX_NIL_REGISTER);
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
if ((ilen+3) < remain) {
More optimization for string instructions
2007-12-17 22:52:01 +03:00
i->modRMForm.displ32u = FetchDWORD(iptr);
Optimizing fetchdecode
2003-08-15 17:08:24 +04:00
iptr += 4;
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
ilen += 4;
}
small optimization
2009-12-21 16:38:06 +03:00
else return(-1);
In 64-bit mode, the CS, DS, ES, and SS segment overrides are ignored.
2005-02-12 22:25:33 +03:00
}
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
// mod==00b, rm!=4, rm!=5
goto modrm_done;
In 64-bit mode, the CS, DS, ES, and SS segment overrides are ignored.
2005-02-12 22:25:33 +03:00
}
decode tables opt
2010-02-01 10:59:22 +03:00
seg = sreg_mod1or2_base32[rm];
In 64-bit mode, the CS, DS, ES, and SS segment overrides are ignored.
2005-02-12 22:25:33 +03:00
}
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
else { // mod!=11b, rm==4, s-i-b byte follows
Updated accessing of modrm/sib addressing information to use accessors. This lets me work on compressing the size of fetch-decode structure (now called bxInstruction_c). I've reduced it down to about 76 bytes. We should be able to do much better soon. I needed the abstraction of the accessors, so I have a lot of freedom to re-arrange things without making massive future changes. Lost a few percent of performance in these mods, but my main focus was to get the abstraction.
2002-09-18 02:50:53 +04:00
unsigned sib, base, index, scale;
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
if (ilen < remain) {
sib = *iptr++;
ilen++;
In 64-bit mode, the CS, DS, ES, and SS segment overrides are ignored.
2005-02-12 22:25:33 +03:00
}
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
else {
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
In 64-bit mode, the CS, DS, ES, and SS segment overrides are ignored.
2005-02-12 22:25:33 +03:00
}
optimize modrm/sib decoding
2007-11-19 00:38:58 +03:00
base = sib & 0x7; sib >>= 3;
index = sib & 0x7; sib >>= 3;
Updated accessing of modrm/sib addressing information to use accessors. This lets me work on compressing the size of fetch-decode structure (now called bxInstruction_c). I've reduced it down to about 76 bytes. We should be able to do much better soon. I needed the abstraction of the accessors, so I have a lot of freedom to re-arrange things without making massive future changes. Lost a few percent of performance in these mods, but my main focus was to get the abstraction.
2002-09-18 02:50:53 +04:00
scale = sib;
Unpack more fields in bxInstruction_c -> this increase bxInstruction size by 4 bytes but I have no way but do it if want to support SSE5 dest override later
2008-02-05 00:28:53 +03:00
i->setSibScale(scale);
Cleanups. Move bxInstruction_c definition to separate file instr.h
2008-01-29 20:13:10 +03:00
i->setSibBase(base);
Reorganize fetchdecode tables with another level of redirection - a leap toward future improvements Currently no speedup and no slowdown - about the same results on my Bochs benchmarking A lot of code reorganization in fetchdecode
2008-04-05 02:39:45 +04:00
if (index != 4) {
More optimization for string instructions
2007-12-17 22:52:01 +03:00
i->ResolveModrm = &BX_CPU_C::BxResolve32BaseIndex;
preparations for future optimizations - not necessary speedupo now
2008-04-03 21:56:59 +04:00
i->setSibIndex(index);
}
Updated accessing of modrm/sib addressing information to use accessors. This lets me work on compressing the size of fetch-decode structure (now called bxInstruction_c). I've reduced it down to about 76 bytes. We should be able to do much better soon. I needed the abstraction of the accessors, so I have a lot of freedom to re-arrange things without making massive future changes. Lost a few percent of performance in these mods, but my main focus was to get the abstraction.
2002-09-18 02:50:53 +04:00
if (mod == 0x00) { // mod==00b, rm==4
optimize seg_override decoding
2008-09-08 20:15:59 +04:00
seg = sreg_mod0_base32[base];
fetchdecode code duplication cleanup
2010-02-06 20:14:07 +03:00
if (base == 5) {
preparations for future optimizations - not necessary speedupo now
2008-04-03 21:56:59 +04:00
i->setSibBase(BX_NIL_REGISTER);
Unroll back 32-bit fetchdecode displ
2008-08-12 01:06:27 +04:00
if ((ilen+3) < remain) {
i->modRMForm.displ32u = FetchDWORD(iptr);
iptr += 4;
ilen += 4;
}
else {
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
Unroll back 32-bit fetchdecode displ
2008-08-12 01:06:27 +04:00
}
Remove BxResolve tables
2007-11-19 01:14:39 +03:00
}
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
// mod==00b, rm==4, base!=5
goto modrm_done;
In 64-bit mode, the CS, DS, ES, and SS segment overrides are ignored.
2005-02-12 22:25:33 +03:00
}
optimize seg_override decoding
2008-09-08 20:15:59 +04:00
seg = sreg_mod1or2_base32[base];
fetchdecode code duplication cleanup
2010-02-06 20:14:07 +03:00
}
if (mod == 0x40) { // mod==01b
if (ilen < remain) {
// 8 sign extended to 32
i->modRMForm.displ32u = (Bit8s) *iptr++;
ilen++;
Unroll back 32-bit fetchdecode displ
2008-08-12 01:06:27 +04:00
goto modrm_done;
}
else {
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
Unroll back 32-bit fetchdecode displ
2008-08-12 01:06:27 +04:00
}
fetchdecode code duplication cleanup
2010-02-06 20:14:07 +03:00
}
// (mod == 0x80), mod==10b
if ((ilen+3) < remain) {
i->modRMForm.displ32u = FetchDWORD(iptr);
iptr += 4;
ilen += 4;
}
else {
return(-1);
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
}
In 64-bit mode, the CS, DS, ES, and SS segment overrides are ignored.
2005-02-12 22:25:33 +03:00
}
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
else {
// 16-bit addressing modes, mod==11b handled above
Hide some BX_MEM_C variables Optimize resolve16 methods - by reducing their amount again - reduce chance for misspredictin
2008-05-02 00:08:37 +04:00
i->ResolveModrm = &BX_CPU_C::BxResolve16BaseIndex;
i->setSibBase(Resolve16BaseReg[rm]);
i->setSibIndex(Resolve16IndexReg[rm]);
optimize modrm/sib decoding
2007-11-19 00:38:58 +03:00
if (mod == 0x00) { // mod == 00b
optimize seg_override decoding
2008-09-08 20:15:59 +04:00
seg = sreg_mod00_rm16[rm];
fetchdecode code duplication cleanup
2010-02-06 20:14:07 +03:00
if (rm == 6) {
Try to optimize individual instructions after fetchdecode
2008-09-16 23:20:03 +04:00
i->setSibBase(BX_NIL_REGISTER);
optimize modrm/sib decoding
2007-11-19 00:38:58 +03:00
if ((ilen+1) < remain) {
More optimization for string instructions
2007-12-17 22:52:01 +03:00
i->modRMForm.displ16u = FetchWORD(iptr);
optimize modrm/sib decoding
2007-11-19 00:38:58 +03:00
iptr += 2;
ilen += 2;
goto modrm_done;
}
small optimization
2009-12-21 16:38:06 +03:00
else return(-1);
optimize modrm/sib decoding
2007-11-19 00:38:58 +03:00
}
goto modrm_done;
}
optimize seg_override decoding
2008-09-08 20:15:59 +04:00
seg = sreg_mod01or10_rm16[rm];
optimize modrm/sib decoding
2007-11-19 00:38:58 +03:00
if (mod == 0x40) { // mod == 01b
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
if (ilen < remain) {
// 8 sign extended to 16
More optimization for string instructions
2007-12-17 22:52:01 +03:00
i->modRMForm.displ16u = (Bit8s) *iptr++;
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
ilen++;
goto modrm_done;
}
fetchdecode code duplication cleanup
2010-02-06 20:14:07 +03:00
else {
return(-1);
}
In 64-bit mode, the CS, DS, ES, and SS segment overrides are ignored.
2005-02-12 22:25:33 +03:00
}
optimize modrm/sib decoding
2007-11-19 00:38:58 +03:00
// (mod == 0x80) mod == 10b
Unroll back 32-bit fetchdecode displ
2008-08-12 01:06:27 +04:00
if ((ilen+1) < remain) {
i->modRMForm.displ16u = FetchWORD(iptr);
iptr += 2;
ilen += 2;
}
fetchdecode code duplication cleanup
2010-02-06 20:14:07 +03:00
else {
return(-1);
}
In 64-bit mode, the CS, DS, ES, and SS segment overrides are ignored.
2005-02-12 22:25:33 +03:00
}
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
modrm_done:
Changes method of resolving opcode/attributes from group table New method more flexible and easy to understanding. Reorganizing fetchdecode code and make it more easy and understandable
2003-08-28 23:25:23 +04:00
// Resolve ExecutePtr and additional opcode Attr
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
const BxOpcodeInfo_t *OpcodeInfoPtr = &(BxOpcodeInfo32[index]);
attr = BxOpcodeInfo32[index].Attr;
split11b on opcode tables level - split almost eevery splittable instruction will be continued
2007-11-17 15:44:10 +03:00
Small code cleanup
2008-03-29 12:58:23 +03:00
while(attr & BxGroupX) {
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
Bit32u group = attr & BxGroupX;
Small code cleanup
2008-03-29 12:58:23 +03:00
attr &= ~BxGroupX;
white space cleanup
2008-02-03 00:46:54 +03:00
a little simpler fetchdecode
2010-05-22 01:17:32 +04:00
if (group < BxPrefixSSE) {
/* For opcodes with only one allowed SSE prefix */
if (sse_prefix != (group >> 4)) {
OpcodeInfoPtr = &BxOpcodeGroupSSE_ERR[0]; // BX_IA_ERROR
}
/* get additional attributes from group table */
attr |= OpcodeInfoPtr->Attr;
break;
}
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
switch(group) {
Small code cleanup
2008-03-29 12:58:23 +03:00
case BxGroupN:
OpcodeInfoPtr = &(OpcodeInfoPtr->AnotherArray[nnn]);
break;
case Bx3ByteOp:
OpcodeInfoPtr = &(OpcodeInfoPtr->AnotherArray[b3]);
break;
Implemented MOVBE Intel Atom(R) instruction
2008-08-11 22:53:24 +04:00
case BxOSizeGrp:
OpcodeInfoPtr = &(OpcodeInfoPtr->AnotherArray[os_32]);
break;
Small code cleanup
2008-03-29 12:58:23 +03:00
case BxPrefixSSE:
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
/* For SSE opcodes look into another table
Small code cleanup
2008-03-29 12:58:23 +03:00
with the opcode prefixes (NONE, 0x66, 0xF2, 0xF3) */
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
if (sse_prefix) {
OpcodeInfoPtr = &(OpcodeInfoPtr->AnotherArray[sse_prefix-1]);
break;
}
continue;
Small code cleanup
2008-03-29 12:58:23 +03:00
case BxFPEscape:
A bit simplify FPU decoding
2008-09-16 22:28:53 +04:00
OpcodeInfoPtr = &(OpcodeInfoPtr->AnotherArray[b2 & 0x3f]);
Small code cleanup
2008-03-29 12:58:23 +03:00
break;
default:
BX_PANIC(("fetchdecode: Unknown opcode group"));
}
Changes method of resolving opcode/attributes from group table New method more flexible and easy to understanding. Reorganizing fetchdecode code and make it more easy and understandable
2003-08-28 23:25:23 +04:00
Small code cleanup
2008-03-29 12:58:23 +03:00
/* get additional attributes from group table */
attr |= OpcodeInfoPtr->Attr;
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
}
Changes method of resolving opcode/attributes from group table New method more flexible and easy to understanding. Reorganizing fetchdecode code and make it more easy and understandable
2003-08-28 23:25:23 +04:00
Inline pop/push functions Store only single byte of opcode in b1() - speedup shift instructions Code cleanups
2008-04-05 21:51:55 +04:00
ia_opcode = OpcodeInfoPtr->IA;
Changes method of resolving opcode/attributes from group table New method more flexible and easy to understanding. Reorganizing fetchdecode code and make it more easy and understandable
2003-08-28 23:25:23 +04:00
}
else {
// Opcode does not require a MODRM byte.
// Note that a 2-byte opcode (0F XX) will jump to before
// the if() above after fetching the 2nd byte, so this path is
// taken in all cases if a modrm byte is NOT required.
Separate PAUSE instruction from regular NOP
2009-01-27 23:29:05 +03:00
fetchdecode tables reorg phase1
2010-05-13 09:38:24 +04:00
const BxOpcodeInfo_t *OpcodeInfoPtr = &(BxOpcodeInfo32[index]);
Separate PAUSE instruction from regular NOP
2009-01-27 23:29:05 +03:00
unsigned group = attr & BxGroupX;
BX_ASSERT(group == BxPrefixSSE || group == 0);
if (group == BxPrefixSSE && sse_prefix)
OpcodeInfoPtr = &(OpcodeInfoPtr->AnotherArray[sse_prefix-1]);
ia_opcode = OpcodeInfoPtr->IA;
Unpack more fields in bxInstruction_c -> this increase bxInstruction size by 4 bytes but I have no way but do it if want to support SSE5 dest override later
2008-02-05 00:28:53 +03:00
i->setOpcodeReg(b1 & 7);
Merged BOCHS-SSE branch
2002-10-16 21:37:35 +04:00
}
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
Read attributes bits even for BxSplit11b opcodes Move lock prefix check later in fetchdecode function when all attributes is ready.
2003-04-06 23:08:31 +04:00
if (lock) { // lock prefix invalid opcode
Simplify and optimize fetchdecode methods. Now fetchdecode is simpler to understand and easier to modify, for example to support 3-byte opcodes (SSE4)
2006-04-06 00:52:40 +04:00
// lock prefix not allowed or destination operand is not memory
Split last BxLockable opcodes -> this allows to eliminate mod==0xc0 check from fetchdecode of every instruction reduce ACPU.CC dependencies - now that file doesn't depend of CPU
2007-11-25 23:22:10 +03:00
// mod == 0xc0 can't be BxLockable in fetchdecode tables
if (/*(mod == 0xc0) ||*/ !(attr & BxLockable)) {
Simplify and optimize fetchdecode methods. Now fetchdecode is simpler to understand and easier to modify, for example to support 3-byte opcodes (SSE4)
2006-04-06 00:52:40 +04:00
BX_INFO(("LOCK prefix unallowed (op1=0x%x, attr=0x%x, mod=0x%x, nnn=%u)", b1, attr, mod, nnn));
Trace cache optimization merged
2007-12-09 21:36:05 +03:00
// replace execution function with undefined-opcode
Inline pop/push functions Store only single byte of opcode in b1() - speedup shift instructions Code cleanups
2008-04-05 21:51:55 +04:00
ia_opcode = BX_IA_ERROR;
Simplify and optimize fetchdecode methods. Now fetchdecode is simpler to understand and easier to modify, for example to support 3-byte opcodes (SSE4)
2006-04-06 00:52:40 +04:00
}
Read attributes bits even for BxSplit11b opcodes Move lock prefix check later in fetchdecode function when all attributes is ready.
2003-04-06 23:08:31 +04:00
}
Fixed lazy flags bug I added in one of my prev merges ICACHE code reorganization
2009-03-23 00:12:35 +03:00
i->modRMForm.Id = 0;
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
imm_mode = attr & BxImmediate;
if (imm_mode) {
switch (imm_mode) {
a bit simplify and optimize shift instructions print failed segment info in check_cs - more debug info
2007-12-30 23:16:35 +03:00
case BxImmediate_I1:
i->modRMForm.Ib = 1;
break;
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
case BxImmediate_Ib:
if (ilen < remain) {
More optimization for string instructions
2007-12-17 22:52:01 +03:00
i->modRMForm.Ib = *iptr;
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
ilen++;
In 64-bit mode, the CS, DS, ES, and SS segment overrides are ignored.
2005-02-12 22:25:33 +03:00
}
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
else {
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
In 64-bit mode, the CS, DS, ES, and SS segment overrides are ignored.
2005-02-12 22:25:33 +03:00
}
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
break;
case BxImmediate_Ib_SE: // Sign extend to OS size
if (ilen < remain) {
Cleanup and optimize
2004-08-28 12:41:46 +04:00
Bit8s temp8s = *iptr;
More optimization for string instructions
2007-12-17 22:52:01 +03:00
if (i->os32L())
i->modRMForm.Id = (Bit32s) temp8s;
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
else
More optimization for string instructions
2007-12-17 22:52:01 +03:00
i->modRMForm.Iw = (Bit16s) temp8s;
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
ilen++;
In 64-bit mode, the CS, DS, ES, and SS segment overrides are ignored.
2005-02-12 22:25:33 +03:00
}
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
else {
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
In 64-bit mode, the CS, DS, ES, and SS segment overrides are ignored.
2005-02-12 22:25:33 +03:00
}
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
break;
Decode SSE4A Rework immediate bytes decoding to make it faster
2007-12-15 20:42:24 +03:00
case BxImmediate_Iw:
if ((ilen+1) < remain) {
More optimization for string instructions
2007-12-17 22:52:01 +03:00
i->modRMForm.Iw = FetchWORD(iptr);
Decode SSE4A Rework immediate bytes decoding to make it faster
2007-12-15 20:42:24 +03:00
ilen += 2;
}
else {
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
Decode SSE4A Rework immediate bytes decoding to make it faster
2007-12-15 20:42:24 +03:00
}
break;
Fixes in resolve.cc -> reduce amount of resolve functions even more
2007-12-19 00:41:44 +03:00
case BxImmediate_Id:
if ((ilen+3) < remain) {
i->modRMForm.Id = FetchDWORD(iptr);
ilen += 4;
}
else {
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
Fixes in resolve.cc -> reduce amount of resolve functions even more
2007-12-19 00:41:44 +03:00
}
break;
Clean invalidate_prefetch_q when not needed
2008-08-23 17:55:37 +04:00
case BxImmediate_BrOff8:
if (ilen < remain) {
i->modRMForm.Id = (Bit8s) (*iptr);
ilen++;
}
else {
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
Clean invalidate_prefetch_q when not needed
2008-08-23 17:55:37 +04:00
}
break;
branch_16 optimizations
2008-08-24 02:27:58 +04:00
case BxImmediate_IwIb:
bugfix + small optimization
2009-11-06 00:07:18 +03:00
if ((ilen+2) < remain) {
i->IxIxForm.Iw = FetchWORD(iptr);
branch_16 optimizations
2008-08-24 02:27:58 +04:00
iptr += 2;
More optimization for string instructions
2007-12-17 22:52:01 +03:00
i->IxIxForm.Ib2 = *iptr;
bugfix + small optimization
2009-11-06 00:07:18 +03:00
ilen += 3;
cleanups and fixes for Immediate_IbIb of SSE4A
2007-12-16 23:47:10 +03:00
}
else {
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
cleanups and fixes for Immediate_IbIb of SSE4A
2007-12-16 23:47:10 +03:00
}
break;
branch_16 optimizations
2008-08-24 02:27:58 +04:00
case BxImmediate_IwIw: // CALL_Ap
bugfix + small optimization
2009-11-06 00:07:18 +03:00
if ((ilen+3) < remain) {
i->IxIxForm.Iw = FetchWORD(iptr);
Decode SSE4A Rework immediate bytes decoding to make it faster
2007-12-15 20:42:24 +03:00
iptr += 2;
branch_16 optimizations
2008-08-24 02:27:58 +04:00
i->IxIxForm.Iw2 = FetchWORD(iptr);
bugfix + small optimization
2009-11-06 00:07:18 +03:00
ilen += 4;
Decode SSE4A Rework immediate bytes decoding to make it faster
2007-12-15 20:42:24 +03:00
}
else {
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
Decode SSE4A Rework immediate bytes decoding to make it faster
2007-12-15 20:42:24 +03:00
}
break;
branch_16 optimizations
2008-08-24 02:27:58 +04:00
case BxImmediate_IdIw: // CALL_Ap
bugfix + small optimization
2009-11-06 00:07:18 +03:00
if ((ilen+5) < remain) {
branch_16 optimizations
2008-08-24 02:27:58 +04:00
i->IxIxForm.Id = FetchDWORD(iptr);
iptr += 4;
More optimization for string instructions
2007-12-17 22:52:01 +03:00
i->IxIxForm.Iw2 = FetchWORD(iptr);
bugfix + small optimization
2009-11-06 00:07:18 +03:00
ilen += 6;
In 64-bit mode, the CS, DS, ES, and SS segment overrides are ignored.
2005-02-12 22:25:33 +03:00
}
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
else {
small optimization
2009-12-21 16:38:06 +03:00
return(-1);
In 64-bit mode, the CS, DS, ES, and SS segment overrides are ignored.
2005-02-12 22:25:33 +03:00
}
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
break;
case BxImmediate_O:
More optimization for string instructions
2007-12-17 22:52:01 +03:00
// For is which embed the address in the opcode.
if (i->as32L()) {
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
// fetch 32bit address into Id
if ((ilen+3) < remain) {
More optimization for string instructions
2007-12-17 22:52:01 +03:00
i->modRMForm.Id = FetchDWORD(iptr);
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
ilen += 4;
}
small optimization
2009-12-21 16:38:06 +03:00
else return(-1);
In 64-bit mode, the CS, DS, ES, and SS segment overrides are ignored.
2005-02-12 22:25:33 +03:00
}
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
else {
// fetch 16bit address into Id
if ((ilen+1) < remain) {
More optimization for string instructions
2007-12-17 22:52:01 +03:00
i->modRMForm.Id = (Bit32u) FetchWORD(iptr);
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
ilen += 2;
}
small optimization
2009-12-21 16:38:06 +03:00
else return(-1);
In 64-bit mode, the CS, DS, ES, and SS segment overrides are ignored.
2005-02-12 22:25:33 +03:00
}
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
break;
default:
Changes method of resolving opcode/attributes from group table New method more flexible and easy to understanding. Reorganizing fetchdecode code and make it more easy and understandable
2003-08-28 23:25:23 +04:00
BX_INFO(("b1 was %x", b1));
strip '\n' from BX_{INFO,DEBUG,ERROR,PANIC} don't need it, moved the output of it into the general io functions. saves space, as well as removes the confusing output if a '\n' is left off
2001-05-30 22:56:02 +04:00
BX_PANIC(("fetchdecode: imm_mode = %u", imm_mode));
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
}
Handle NULL_SEG_REG (no segment override) case in fetchdecode.cc
2006-05-25 00:57:37 +04:00
}
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
Merged NEW-INSTRUCTIONS branch
2003-05-15 20:41:17 +04:00
#if BX_SUPPORT_3DNOW
Handle NULL_SEG_REG (no segment override) case in fetchdecode.cc
2006-05-25 00:57:37 +04:00
if(b1 == 0x10f)
optimized decode tables
2010-01-31 12:45:27 +03:00
ia_opcode = Bx3DNowOpcode[i->modRMForm.Ib];
Merged NEW-INSTRUCTIONS branch
2003-05-15 20:41:17 +04:00
#endif
optimize seg_override decoding
2008-09-08 20:15:59 +04:00
if (! BX_NULL_SEG_REG(seg_override))
seg = seg_override;
i->setSeg(seg);
Handle NULL_SEG_REG (no segment override) case in fetchdecode.cc
2006-05-25 00:57:37 +04:00
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
if (attr & BxArithDstRM) {
Clear some duplicated arithmetic opcodes - difference only in operands order
2008-07-13 13:59:59 +04:00
i->setRm(nnn);
i->setNnn(rm);
}
Some kind of big change in fetchdecode tables invented in order to compress the tables for better host data cache utilization
2008-08-29 23:23:03 +04:00
i->execute = BxOpcodesTable[ia_opcode].execute1;
i->execute2 = BxOpcodesTable[ia_opcode].execute2;
More optimization for string instructions
2007-12-17 22:52:01 +03:00
i->setB1(b1);
i->setILen(ilen);
added ia_opcode into bxInstruction class
2010-01-09 18:11:32 +03:00
i->setIaOpcode(ia_opcode);
Instrumentation code updated, some PANIC messages fixed
2008-06-23 06:56:31 +04:00
small optimization
2009-12-21 16:38:06 +03:00
#if BX_SUPPORT_TRACE_CACHE
if ((attr & BxTraceEnd) || ia_opcode == BX_IA_ERROR)
return(1);
#endif
return(0);
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
}
Use fastcall gcc attribute for all cpu execution functions - this pure "compiler helper" optimization brings additional 2% speedup to Bochs code
2008-03-23 00:29:41 +03:00
void BX_CPP_AttrRegparmN(1) BX_CPU_C::BxError(bxInstruction_c *i)
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
{
allow to configure CPU features at runtime - implemened on example of SSE/AES/MOVBE/POPCNT
2010-02-26 01:04:31 +03:00
unsigned ia_opcode = i->getIaOpcode();
if (ia_opcode == BX_IA_ERROR) {
BX_DEBUG(("BxError: Encountered an unknown instruction b1=0x%02x (signalling #UD)", i->b1()));
BX_DEBUG(("modrm was 0x%02x, nnn was %u, rm was %u", i->modrm(), i->nnn(), i->rm()));
This more ellegant way to have debug info for BxError and not lose any performace
2008-08-09 23:18:09 +04:00
#if BX_DISASM && BX_DEBUGGER == 0 // with debugger it easy to see the #UD
allow to configure CPU features at runtime - implemened on example of SSE/AES/MOVBE/POPCNT
2010-02-26 01:04:31 +03:00
if (LOG_THIS getonoff(LOGLEV_DEBUG))
debug_disasm_instruction(BX_CPU_THIS_PTR prev_rip);
This more ellegant way to have debug info for BxError and not lose any performace
2008-08-09 23:18:09 +04:00
#endif
allow to configure CPU features at runtime - implemened on example of SSE/AES/MOVBE/POPCNT
2010-02-26 01:04:31 +03:00
}
else {
enable compilation with CPU_LEVEL <= 6 converted SEP to runtime option as well
2010-02-26 14:44:50 +03:00
BX_INFO(("%s: instruction not supported - signalling #UD (features bitmask: 0x%08x)",
changes variable name to better one
2010-04-03 09:59:07 +04:00
get_bx_opcode_name(ia_opcode), BX_CPU_THIS_PTR isa_extensions_bitmask));
allow to configure CPU features at runtime - implemened on example of SSE/AES/MOVBE/POPCNT
2010-02-26 01:04:31 +03:00
}
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
remove unused param from exception() call
2010-03-14 18:51:27 +03:00
exception(BX_UD_EXCEPTION, 0);
- entered original Bochs snapshot bochs-2000_0325a.tar.gz from ftp.bochs.com
2001-04-10 05:04:59 +04:00
}
added ia_opcode into bxInstruction class
2010-01-09 18:11:32 +03:00
const char *get_bx_opcode_name(Bit16u ia_opcode)
{
static const char* BxOpcodeNamesTable[BX_IA_LAST] =
{
1st step of moving CPU options to runtime
2010-02-24 22:27:51 +03:00
#define bx_define_opcode(a, b, c, d) #a,
added ia_opcode into bxInstruction class
2010-01-09 18:11:32 +03:00
#include "ia_opcodes.h"
};
#undef bx_define_opcode
return (ia_opcode < BX_IA_LAST) ? BxOpcodeNamesTable[ia_opcode] : 0;
}
allow to configure CPU features at runtime - implemened on example of SSE/AES/MOVBE/POPCNT
2010-02-26 01:04:31 +03:00
void BX_CPU_C::init_FetchDecodeTables(void)
{
static Bit32u BxOpcodeFeatures[BX_IA_LAST] =
{
#define bx_define_opcode(a, b, c, d) d,
#include "ia_opcodes.h"
};
#undef bx_define_opcode
changes variable name to better one
2010-04-03 09:59:07 +04:00
Bit32u features = BX_CPU_THIS_PTR isa_extensions_bitmask;
fix PANIC on 386
2010-04-07 20:02:52 +04:00
#if BX_CPU_LEVEL > 3
allow to configure CPU features at runtime - implemened on example of SSE/AES/MOVBE/POPCNT
2010-02-26 01:04:31 +03:00
if (! features)
BX_PANIC(("init_FetchDecodeTables: CPU features bitmask is empty !"));
fix PANIC on 386
2010-04-07 20:02:52 +04:00
#endif
fixes
2010-03-05 22:49:22 +03:00
if (BX_IA_LAST > 0xffff)
BX_PANIC(("init_FetchDecodeTables: too many opcodes defined !"));
allow to configure CPU features at runtime - implemened on example of SSE/AES/MOVBE/POPCNT
2010-02-26 01:04:31 +03:00
for (unsigned n=0; n < BX_IA_LAST; n++) {
Bit32u ia_opcode_features = BxOpcodeFeatures[n];
if (ia_opcode_features) {
if ((ia_opcode_features & features) == 0)
BxOpcodesTable[n].execute1 = &BX_CPU_C::BxError;
}
}
}
Reference in New Issue Copy Permalink