Commit Graph

79 Commits

Author SHA1 Message Date
Stanislav Shwartsman
2b5a812674 Split last bit.cc methods according to os16/32/64 2005-07-25 04:18:20 +00:00
Stanislav Shwartsman
ce8f1ade07 Some not really significant speedups 2005-06-21 17:01:21 +00:00
Stanislav Shwartsman
015ad92958 Added SMP status to TODO file
Removed abusive BX_INFO from WBINVD instruction
The PREFETCHW (3DNow!) instruction should not #UD in x86-64 even on Intel w/o 3DNow!
2005-05-27 01:53:38 +00:00
Stanislav Shwartsman
663f7d5ef3 CMPXCHG16B instruction implemented 2005-05-19 20:25:16 +00:00
Stanislav Shwartsman
a86002a8bc Improve Bochs instrumentation
Small changes in APIC timer, should fix the bug report
[ 957660 ] >>PANIC<< APIC: R(curr timer count): delta < initial
2005-04-29 21:28:59 +00:00
Stanislav Shwartsman
619942cf9a Enable SYSENTER/SYSEXIT together with x86-64 support, these instructions used by gentoo amd64 LiveCD image (at least it WRMSR to SYSENTER MSRs).
SYSENTER/SYSEXIT is not recognized in long mode but it could be used i any other mode without problem
2005-03-29 21:59:44 +00:00
Stanislav Shwartsman
e6e9dd3825 Extend Bochs instrumentation
Compatability fixes
2005-03-17 20:50:57 +00:00
Stanislav Shwartsman
709b218c10 Reduce metaInfo initialization in fetchDecode 2005-03-01 21:44:01 +00:00
Stanislav Shwartsman
2bfc842c09 CPU fixes by Kevin Lawton 2005-02-16 21:27:21 +00:00
Stanislav Shwartsman
9492942ae6 In 64-bit mode, the CS, DS, ES, and SS segment overrides are ignored. 2005-02-12 19:25:33 +00:00
Stanislav Shwartsman
f375203fdb preparations for x86-64 support in disasm 2004-12-12 22:12:43 +00:00
Stanislav Shwartsman
46bb3d8853 remove duplicated data arrays from CPU 2004-12-11 20:51:13 +00:00
Stanislav Shwartsman
5213e903bd mov duplicate opcode groups from fectchdecode*.cc to .h
use common register accessor macroses instead of direct register file structure access
2004-11-26 20:21:28 +00:00
Stanislav Shwartsman
69c0b06955 fixes in disassembler
split REPEAT instructions according to opsize to speedup execution
now each REPEATABLE instruction splitted to 3 different instructions, one for 16-bit operand size, one for 32-bit and one for 64-bit. Choosing of correct instruction occure in fetchdecode step.
2004-11-20 23:26:32 +00:00
Stanislav Shwartsman
730b8c0243 Fix this pointers in the code 2004-11-14 21:25:42 +00:00
Stanislav Shwartsman
4f1f070c37 Fix comments for code 2004-10-08 19:29:04 +00:00
Stanislav Shwartsman
3adc5c8659 Fix lock prefix for XOR instruction in 64-bit mode 2004-10-08 19:07:18 +00:00
Stanislav Shwartsman
760a195c9d * Fix LOCK prefix handling for x86-64
* Split BT*_EvGv functions to 3 different function according to exec mode
2004-09-17 20:47:19 +00:00
Stanislav Shwartsman
fc631037ff remove obsolete comments from fetchdecode 2004-09-06 20:22:39 +00:00
Stanislav Shwartsman
279d207d45 Fix fetchdecode bugs reported by Gilbert Netzer
(opcode patches for x86_64 cpu)
2004-05-03 17:58:36 +00:00
Stanislav Shwartsman
0eb71999db Added missed 287 opcodes which should be executed as NOP in 387+ 2003-12-28 18:19:41 +00:00
Stanislav Shwartsman
9ccb363ec3 bochs style decode/execute of FPU instructions.
With this coding style each instruction could be implemented separatelly even not together with current Bochs FPU emulator.
Step-by-step I am going to transfer all FPU instructions from current Bochs FPU emulator to new style and remove an old bugged emulator.
Anyway, now I could implement all currently missed FPU instructions without hacking wm-fpu-emu.
2003-12-27 13:50:06 +00:00
Stanislav Shwartsman
ac20b6405a - FXSAVE/FXRSTOR instructions should be available in P6 mode
- Added second UD2 opcode to fetchdecode
- Added RDPMC instruction to fetchdecode
- 'changes' updated
2003-10-24 18:34:16 +00:00
Stanislav Shwartsman
7f570b0150 Added PNI new streaming extensions instructions
PNI could be enabled by setting BX_SUPPORT_PNI in config.h
After the feature will be fully validation I'll also add configure option.

The implemntation is ~complete. I've missed only three FPU new opcodes of FUSTTP instruction and MONITOR/WAIT instructions.

Enjoy ! ;)
2003-08-29 21:20:52 +00:00
Stanislav Shwartsman
254ad17328 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 19:25:23 +00:00
Stanislav Shwartsman
f6711d51f2 Fixed very serious bug in x86-64 with wrong decoding of opcodes with two immediates 2003-08-15 13:17:16 +00:00
Stanislav Shwartsman
3c00944998 I hope this is the last one ... 2003-05-29 19:44:59 +00:00
Stanislav Shwartsman
56cc6469e3 Ops, missed one ... 2003-05-29 17:19:38 +00:00
Stanislav Shwartsman
f933d604d3 Fixed missed BxLockable for XCHG instruction 2003-05-29 17:15:08 +00:00
Stanislav Shwartsman
140ff2541b Ability to configure x86-64 without 3DNOW 2003-05-21 15:20:51 +00:00
Stanislav Shwartsman
1d45167e5b Merged NEW-INSTRUCTIONS branch 2003-05-15 16:41:17 +00:00
Christophe Bothamy
83b8bbedff - fix REX MOVB immediate for x86_64 (patch by Arnd Bergmann)
(bug [ 720776 ] REX MOVB immediate broken for x86_64)
2003-04-26 10:02:03 +00:00
Stanislav Shwartsman
446fca9ed0 Superfluous braces in initializers in fetchdecode.cc 2003-04-23 17:52:59 +00:00
Stanislav Shwartsman
40bd4f138b Little style changes
Elliminated i387_t alimit field (not used in FPU)
2003-04-16 18:38:53 +00:00
Stanislav Shwartsman
1d54caca9b Fixed compilation error 2003-04-09 19:20:05 +00:00
Stanislav Shwartsman
7db893970c Read attributes bits even for BxSplit11b opcodes
Move lock prefix check later in fetchdecode function when all attributes is ready.
2003-04-06 19:08:31 +00:00
Stanislav Shwartsman
1e71c9e56e 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 12:16:53 +00:00
Christophe Bothamy
1a518b81fe - add __attribute__((regparm(X))) performance trick with gcc on x86
on some cpu instructions (patch from Conn Clark)
- performance improvement is 1% on win95 boot
2003-03-17 00:41:01 +00:00
Peter Tattam
2f9088a223 x86-64 emulation.
Fixed IRETD in 64 bit mode
2003-03-13 00:45:44 +00:00
Stanislav Shwartsman
cdfc3cbce4 instrumentation enchancements:
* renamed CPU_ID to BX_CPU_ID.
  with this new name there is no possibility for name contentions and BX_CPU_ID
  definition could be moved out to NEED_CPU_REG_SHORTCUTS block

* returned back `unsigned BX_CPU::which_cpu(void)` function

* added BX_CPU_ID parameter for
	BX_INSTR_PHY_READ(a20addr, len);
	BX_INSTR_PHY_WRITE(a20addr, len);
    now it will be
	BX_INSTR_PHY_READ(cpu_id, a20addr, len);
	BX_INSTR_PHY_WRITE(cpu_id, a20addr, len);
2003-02-13 15:04:11 +00:00
Peter Tattam
22d855a6c0 Fixed wrong RETF instructions for 64 bit mode. 2003-02-08 05:51:38 +00:00
Stanislav Shwartsman
4b59ecbc62 Implemented SSE/SSE2 duplicate opcodes in more intellegent way ... 2002-12-22 21:48:23 +00:00
Stanislav Shwartsman
e73df72525 implementation of additional SSE/SSE2 instructions 2002-12-22 20:42:56 +00:00
Stanislav Shwartsman
4906ffef7c Clean Peter's commit with MOVNTDQ instruction implementation 2002-12-20 09:11:39 +00:00
Peter Tattam
c173034663 Patches to round off the x86-64 emulation to get the Linux x86-64 kernel and
sash to run.

1) fixed fetchdecode64.cc to fix the operand size at 64 bits in long mode for moves
   to/from CRx

2) minor patches to sse2.cc to fix unimplemented and 64 bit variants of sse2
instructions.
2002-12-20 07:11:29 +00:00
Bryce Denney
9b2914fd1d - Temporarily revert Stanislav's changes between 2002-12-18 and 2002-12-19.
Because source files were added/removed it would require an update
  of the windows and macos project files, so I want to wait until after 2.0.
    M Makefile.in         1.51 back to 1.50
    M cpu.h               1.121 back to 1.120
    M fetchdecode.cc      1.37 back to 1.36
    M fetchdecode64.cc    1.33 back to 1.32
    M sse.cc              1.17 back to 1.16
    A sse2.cc             1.27 back to 1.26  (added back)
    R sse_move.cc         removed
    R sse_pfp.cc          removed
- to bring these changes back again, all we have to do is
  "cvs update -j tmp-before1 -j tmp-after1"
2002-12-19 05:53:18 +00:00
Stanislav Shwartsman
aa361badf2 Reorganized SSE/SSE2 code
sse.cc -> general SSE stuff and SSE integer (MMX extensions)
sse_move.cc -> memory transfer and shuffle opcodes
sse_pfp.cc -> packed floating point operations
2002-12-18 22:33:44 +00:00
Stanislav Shwartsman
bcd57bdcaf *** Current duplicate SSE/SSE2 instructions list ***
MOVUPS_VpsWps  (0f 10) = MOVUPD_VpdWpd  (66 0f 10) = MOVDQU_VdqWdq (f3 0f 6f)
MOVUPS_WpsVps  (0f 11) = MOVUPD_WpdVpd  (66 0f 11) = MOVDQU_WdqVdq (f3 0f 7f)
MOVAPS_VpsWps  (0f 28) = MOVAPD_VpdWpd  (66 0f 28) = MOVDQA_VdqWdq (66 0f 6f)
MOVAPS_WpsVps  (0f 29) = MOVAPD_WpdVpd  (66 0f 29) = MOVDQA_WdqVdq (66 0f 7f)

MOVNTPS_MdqVps (0f 2b) = MOVNTPD_MdqVpd (66 0f 2b)
MOVLPS_VpsMq   (0f 12) = MOVLPD_VsdMq   (66 0f 12)
MOVLPS_MqVps   (0f 13) = MOVLPD_MqVsd   (66 0f 13)
MOVHPS_VpsMq   (0f 16) = MOVHPD_VpdMq   (66 0f 16)
MOVHPS_MqVps   (0f 17) = MOVHPD_MqVpd   (66 0f 17)

ANDPS_VpsWps   (0f 54) = ANDPD_VpdWpd   (66 0f 54) = PAND_VpdWpd   (66 0f db)
ANDNPS_VpsWps  (0f 55) = ANDNPD_VpdWpd  (66 0f 55) = PANDN_VpdWpd  (66 0f df)
ORPS_VpsWps    (0f 56) = ORPD_VpdWpd    (66 0f 56) = POR_VpdWpd    (66 0f eb)
XORPS_VpsWps   (0f 57) = XORPD_VpdWpd   (66 0f 57) = PXOR_VpdWpd   (66 0f ef)

Removed dupes
2002-11-25 21:58:55 +00:00
Bryce Denney
0a7cb3a43c - apply patch.ifdef-disabled-options. Comments from that patch are below:
For a whole lot of configure options, I put #if...#endif around code that
  is specific to the option, even in files which are normally only compiled
  when the option is on.  This allows me to create a MS Visual C++ 6.0
  workspace that supports many of these options.  The workspace will basically
  compile every file all the time, but the code for disabled options will
  be commented out by the #if...#endif.

  This may one day lead to simplification of the Makefiles and configure
  scripts, but for the moment I'm leaving Makefiles and configure scripts
  alone.

  Affected options:
    BX_SUPPORT_APIC (cpu/apic.cc)
    BX_SUPPORT_X86_64 (cpu/*64.cc)
    BX_DEBUGGER (debug/*)
    BX_DISASM (disasm/*)
    BX_WITH_nameofgui (gui/*)
    BX_SUPPORT_CDROM (iodev/cdrom.cc)
    BX_NE2K_SUPPORT (iodev/eth*.cc, iodev/ne2k.cc)
    BX_SUPPORT_APIC (iodev/ioapic.cc)
    BX_IODEBUG_SUPPORT (iodev/iodebug.cc)
    BX_PCI_SUPPORT (iodev/pci*.cc)
    BX_SUPPORT_SB16 (iodev/sb*.cc)

Modified Files:
  cpu/apic.cc cpu/arith64.cc cpu/ctrl_xfer64.cc
  cpu/data_xfer64.cc cpu/fetchdecode64.cc cpu/logical64.cc
  cpu/mult64.cc cpu/resolve64.cc cpu/shift64.cc cpu/stack64.cc
  debug/Makefile.in debug/crc.cc debug/dbg_main.cc debug/lexer.l
  debug/linux.cc debug/parser.c debug/parser.y
  disasm/dis_decode.cc disasm/dis_groups.cc gui/amigaos.cc
  gui/beos.cc gui/carbon.cc gui/macintosh.cc gui/rfb.cc
  gui/sdl.cc gui/term.cc gui/win32.cc gui/wx.cc gui/wxdialog.cc
  gui/wxmain.cc gui/x.cc iodev/cdrom.cc iodev/eth.cc
  iodev/eth_arpback.cc iodev/eth_fbsd.cc iodev/eth_linux.cc
  iodev/eth_null.cc iodev/eth_packetmaker.cc iodev/eth_tap.cc
  iodev/eth_tuntap.cc iodev/eth_win32.cc iodev/ioapic.cc
  iodev/iodebug.cc iodev/ne2k.cc iodev/pci.cc iodev/pci2isa.cc
  iodev/sb16.cc iodev/soundlnx.cc iodev/soundwin.cc
2002-11-19 05:47:45 +00:00
Stanislav Shwartsman
ccbc8e0ef7 MOVAPS/MOVAPD have a different exceptions 2002-11-15 12:44:39 +00:00