DOS ain't dead

The Bochs Project released version 2.4.2 on 12 November 2009.

Changes:
- CPU and internal debugger
  - VMX: Implemented TPR shadow VMEXIT
  - Bugfixes for CPU emulation correctness (mostly for VMX support).
  - Bugfixes and updates for Bochs internal debugger
    - On SMP system stepN command now affects only current processor

- Memory
  - Bugfixes for > 32-bit physical address space.
  - Allow to emulate more physical memory than host actually could or would
    like to allocate. For more details look for new .bochsrc 'memory' option.

- Cleanup configure options
  - All paging related options now will be automatically determined according
    to  --enable-cpu-level option. Related configure options
        --enable-global-pages, --enable-large-pages,
        --enable-pae, --enable-mtrr
    are deprecated now. Only 1G paging option still remaining unchanged.
  - Deprecate --enable-daz configure option. Denormals-are-zeros MXCSR control
    will be enabled automatically iff SSE2 is supported (like in hardware).
  - Deprecate --enable-vme configure option, now it will be supported iff
    CPU_LEVEL >= 5 (like in hardware).

- I/O Devices
  - Bugfixes for 8254 PIT, VGA, Cirrus-Logic SVGA, USB UCHI

- SF patches applied
 [2817840] Make old_callback static by Mark Marshall
 [2874004] fix for VMWRITE instruction by Roberto Paleari
 [2873999] fix CS segment type during fast syscall invocation by Roberto Paleari
 [2864389] Debugger gui maximize on startup by Thomas Nilsen
 [2817868] Rework loops in the memory code by Mark Marshall
 [2812948] PIT bug by Derek

- these S.F. bugs were closed/fixed
 [2833504] GUI debugger bug-about GDT display
 [2872244] BIOS writes not allowed value to MTRR MSR causing #GP
 [2885383] SDL GUI memory leak
 [2872290] compilation in AIX5.3 ML10 failes
 [2867904] crash with cirrus bx_vga_c::mem_write
 [2851495] BIOS PCI returns with INT flag = 0
 [2860333] vista 64 guest STOP 109 (GDT modification)
 [2849745] disassembler bug for 3DNow and SSE opcodes
 [1066748] Wrong registers values after #RESET, #INIT
 [2836893] Regression: Windows XP installer unable to format harddrive
 [2812239] VMX: VM-Exit: Incorrect instruction length on software int
 [2814130] bx_debug lex/yacc files incorrectly generated
 [2813199] MP Tables Missing From BIOS
 [2824093] VMX exception bug
 [2811909] VMX : CS Access-rights Type.Accessed stays 0
 [2810571] Compile Errors on OSX
 [2823749] GCC regression or VM_EXIT RDMSR/WRMSR bug
 [2815929] Vista/XP64 unnecessary panic
 [2803519] Wrong example in man page bochsrc

- these S.F. feature requests were closed/implemented
  [422766] Large Memory configurations
 [1311287] Idea for a better GUI
  [455971] USB support
  [615363] debugger shortcut for repeat last cmd

Home page: http://sourceforge.net/projects/bochs/
Official announcement: Bochs 2.4.2 released
Download: http://sourceforge.net/projects/bochs/files/

---
Forum admin

> The Bochs Project released version 2.4.2 on 12 November 2009

Could be good news, but ...

> - VMX: Implemented TPR shadow VMEXIT
> - Bugfixes for CPU emulation correctness (mostly for VMX support).

Not that hot for DOS users ...

> - Bugfixes for 8254 PIT, VGA, Cirrus-Logic SVGA, USB UCHI

IIRC Jayson Burgon whined about some "PIT-BUG" in both BOCHS and QEMU some (long) time ago

> [2874004] fix for VMWRITE instruction by Roberto Paleari

http://sf.net/tracker/?func=detail&aid=2874004&group_id=12580&atid=312580

heh

> [2812948] PIT bug by Derek

http://sf.net/tracker/index.php?func=detail&aid=2812948&group_id=12580&atid=312580

> Note that this bug was introduced in 2.4.1, and was not present in 2.3.7.

heh

> [2823749] GCC regression or VM_EXIT RDMSR/WRMSR bug

http://sf.net/tracker/?func=detail&aid=2823749&group_id=12580&atid=112580

This one apparently broke it: it crashes just before booting should start, a deadloop of TripleFault's in WRMSR instruction

2.4 doesn't have the WRMSR bug and still works for me. Also, both 2.4 and 2.4.2 have one useless missing import.

---
This is a LOGITECH mouse driver, but some software expect here
the following string:*** This is Copyright 1983 Microsoft ***

> > [2823749] GCC regression or VM_EXIT RDMSR/WRMSR bug
>
> http://sf.net/tracker/?func=detail&aid=2823749&group_id=12580&atid=112580
>
> This one apparently broke it: it crashes just before booting should start,
> a deadloop of TripleFault's in WRMSR instruction

Bochs 2.4.2 doesn't crash here on Windows XP Professional SP3.

---
Forum admin

In my case all bochs 2.4.x crash on exit :( (it seems doesn't depend on VM running inside, at also happen with bundled linux VM). So I still use old 2.3.6...

---
DOS gives me freedom to unlimited HW access.

rr wrote:

> Bochs 2.4.2 doesn't crash here on Windows XP Professional SP3.

Funny ...

Rayer wrote:

> In my case all bochs 2.4.x crash on exit

For me it crashes just at start, there is no exit, just reboots to the infinity

Someone else can test please ? I tested in DOS only so far.

---
This is a LOGITECH mouse driver, but some software expect here
the following string:*** This is Copyright 1983 Microsoft ***

> For me it crashes just at start, there is no exit, just reboots to the
> infinity
>
> Someone else can test please ? I tested in DOS only so far.

You mean only under HX? Then maybe some win32 api function is not implemented yet. But in my case it crashes under WXP-SP3-CZ.

---
DOS gives me freedom to unlimited HW access.

> You mean only under HX? Then maybe some win32 api function is not
> implemented yet.

Or many ...

> But in my case it crashes under WXP-SP3-CZ.

Well, I apparently had referenced the wrong BIOS from 2.4, so it crashed, with the correct BIOS, it's fine

Still funny that:

- BIOS runs in PM
- Has not even a valid IDCT
- And plays with such dangerous stuff like WRMSR

00008294813i[APIC0] allocate APIC id=0 (MMIO enabled) to 0xfee00000
00008294813i[     ] reset of 'unmapped' plugin device by virtual method
00008294813i[     ] reset of 'biosdev' plugin device by virtual method
00008294813i[     ] reset of 'speaker' plugin device by virtual method
00008294813i[     ] reset of 'extfpuirq' plugin device by virtual method
00008294813i[     ] reset of 'gameport' plugin device by virtual method
00008294813i[     ] reset of 'pci_ide' plugin device by virtual method
00008294813i[     ] reset of 'acpi' plugin device by virtual method
00008294813i[     ] reset of 'ioapic' plugin device by virtual method
00008294813i[     ] reset of 'keyboard' plugin device by virtual method
00008294813i[     ] reset of 'harddrv' plugin device by virtual method
00008294813i[     ] reset of 'serial' plugin device by virtual method
00008294813i[     ] reset of 'parallel' plugin device by virtual method
00008298119i[BIOS ] $Revision: 1.231 $ $Date: 2009/04/26 17:17:07 $
00008612038i[KBD  ] reset-disable command received
00008737808i[VBIOS] VGABios $Id: vgabios.c,v 1.69 2009/04/07 18:18:20 vruppert Exp $
00008737879i[CLVGA] VBE known Display Interface b0c0
00008737911i[CLVGA] VBE known Display Interface b0c5
00008740836i[VBIOS] VBE Bios $Id: vbe.c,v 1.62 2009/01/25 15:46:25 vruppert Exp $
00009047397i[BIOS ] Starting rombios32
00009047894i[BIOS ] Shutdown flag 0
00009048575i[BIOS ] ram_size=0x02000000
00009049053i[BIOS ] ram_end=32MB
00009049305e[CPU0 ] WRMSR[0x00000201]: attempt to write invalid phy addr to variable range MTRR ffffffff:e0000800
00009049305e[CPU0 ] interrupt(): vector must be within IDT table limits, IDT.limit = 0x0
00009049305e[CPU0 ] interrupt(): vector must be within IDT table limits, IDT.limit = 0x0
00009049305i[CPU0 ] CPU is in protected mode (active)
00009049305i[CPU0 ] CS.d_b = 32 bit
00009049305i[CPU0 ] SS.d_b = 32 bit
00009049305i[CPU0 ] EFER   = 0x00000000
00009049305i[CPU0 ] | RAX=00000000e0000800  RBX=00000000000005a0
00009049305i[CPU0 ] | RCX=0000000000000201  RDX=00000000ffffffff
00009049305i[CPU0 ] | RSP=000000000009fb94  RBP=000000000009fbb0
00009049305i[CPU0 ] | RSI=000000000000051c  RDI=0000000002000000
00009049305i[CPU0 ] |  R8=0000000000000000   R9=0000000000000000
00009049305i[CPU0 ] | R10=0000000000000000  R11=0000000000000000
00009049305i[CPU0 ] | R12=0000000000000000  R13=0000000000000000
00009049305i[CPU0 ] | R14=0000000000000000  R15=0000000000000000
00009049305i[CPU0 ] | IOPL=0 id vip vif ac vm RF nt of df if tf sf ZF af PF cf
00009049305i[CPU0 ] | SEG selector     base    limit G D
00009049305i[CPU0 ] | SEG sltr(index|ti|rpl)     base    limit G D
00009049305i[CPU0 ] |  CS:0010( 0002| 0|  0) 00000000 ffffffff 1 1
00009049305i[CPU0 ] |  DS:0018( 0003| 0|  0) 00000000 ffffffff 1 1
00009049305i[CPU0 ] |  SS:0018( 0003| 0|  0) 00000000 ffffffff 1 1
00009049305i[CPU0 ] |  ES:0018( 0003| 0|  0) 00000000 ffffffff 1 1
00009049305i[CPU0 ] |  FS:0000( 0000| 0|  0) 00000000 00000000 0 0
00009049305i[CPU0 ] |  GS:0000( 0000| 0|  0) 00000000 00000000 0 0
00009049305i[CPU0 ] |  MSR_FS_BASE:0000000000000000
00009049305i[CPU0 ] |  MSR_GS_BASE:0000000000000000
00009049305i[CPU0 ] | RIP=00000000000e08a5 (00000000000e08a5)
00009049305i[CPU0 ] | CR0=0x60000011 CR2=0x0000000000000000
00009049305i[CPU0 ] | CR3=0x00000000 CR4=0x00000000
00009049305i[CPU0 ] 0x00000000000e08a5>> wrmsr  : 0F30
00009049305e[CPU0 ] exception(): 3rd (13) exception with no resolution, shutdown status is 00h, resetting
00009049305i[SYS  ] bx_pc_system_c::Reset(HARDWARE) called
00009049305i[CPU0 ] cpu hardware reset
00009049305i[APIC0] allocate APIC id=0 (MMIO enabled) to 0xfee00000

But I have trouble (= hangs) with the mouse wheel in 2.4 and 2.4.2, better (?) in 2.3.7 or without "/O" :

00037480137i[KBD  ] wheel mouse mode request rejected
00041638810i[FDD  ] controller reset in software
00045703313i[FDD  ] controller reset in software
00049767771i[FDD  ] controller reset in software
00053832274i[FDD  ] controller reset in software
00057896777i[FDD  ] controller reset in software
00061961248i[FDD  ] controller reset in software
00066025731i[FDD  ] controller reset in software
00070090205i[FDD  ] controller reset in software
00074154675i[FDD  ] controller reset in software
00078219193i[FDD  ] controller reset in software
00092829357i[FDD  ] controller reset in software
00096893815i[FDD  ] controller reset in software
00100958318i[FDD  ] controller reset in software
00105022776i[FDD  ] controller reset in software
00109087279i[FDD  ] controller reset in software
00113151745i[FDD  ] controller reset in software
00117216248i[FDD  ] controller reset in software
00121280706i[FDD  ] controller reset in software
00125345209i[FDD  ] controller reset in software
00129409667i[FDD  ] controller reset in software
00133474187i[FDD  ] controller reset in software
00137538645i[FDD  ] controller reset in software
00141603148i[FDD  ] controller reset in software
00145667606i[FDD  ] controller reset in software
00149732109i[FDD  ] controller reset in software
00153796575i[FDD  ] controller reset in software
00157861078i[FDD  ] controller reset in software
00161925536i[FDD  ] controller reset in software
00165990039i[FDD  ] controller reset in software
00170054497i[FDD  ] controller reset in software

This ^^^ is from DOS, in XP it also hangs and additionally whines about "keyboard buffer full"

---
This is a LOGITECH mouse driver, but some software expect here
the following string:*** This is Copyright 1983 Microsoft ***