View Full Version : disasseble for arm-processor...
SpeKKeL
March 6th, 2005, 07:59
I've bought a pda with arm processor.
Should i see more
I know ida can disassemble but know i've to learn those arm mnemonics ???
(i doesn't see jumps or calls,sjit!)
So i am looking the best way to reverse the progs of this machine...
Debugging on my xp machine is possible ???
Somebody here who did it earlier?
Yep i read the forum-threads about wince but i need to know whats to best way to start and using what tools.
SpeKK.
naides
March 6th, 2005, 10:26
Try http://Ka0s.net and the links in there.
They have lists of ARM opcodes, links and tuts for debuggers etc.
disavowed
March 6th, 2005, 13:31
In IDA, you can go to Options -> General -> Disassembly tab, and check Auto comments. This may give you enough info on the mnemonics to do your task.
SpeKKeL
March 6th, 2005, 17:31
Thanks for the link and yes that auto comments in ida is very usefull!
Now looking futher how to debug.
Thanks again,
Spekk
SpeKKeL
March 23rd, 2005, 03:56
Hajo,
Trying to debug a pocketpc prog. i followed a tut by hexc ( www.woodmann.com/fravia/hexc_evtdebug.htm ) after installing ms embedded c++ 4.0 i copied the file target file from pocket to my desktop pc> then went to settings and gave in debug tab the download directory from the target of the pocketpc.Then when start to debug all went fine till it tries to copy the file from my desktop comp. :error

ownloading failed while copying the file from my hd... .
Anyone knows what i am doing wrong ??
Thanks,
Spekk
Shub-nigurrath
March 23rd, 2005, 17:47
just yesterday we released an first tut on wince patching, http://tutorials.accessroot.com/ might help you.
SpeKKeL
March 24th, 2005, 07:20
Thanks Shub,
Verry good tutorial (there are so few ).
It seems i had the wrong version or missed some sdk?????
I installed 3.0 and all goes fine.
Back to "Hello world" ans study your tut!
Spekk.
naides
March 24th, 2005, 08:50
That is a very nice and thorough tutorial indeed Shub.
I have one comment (irrelavant to the contents of the tut, just a curosity)
When you state that ARM asm does not have an op code equivalent to the x86 'nop': 0x90, so we have to implement an operation that does nothing, like (MOV R1, R1).
Actually, that is exactly what x86 assembler does: the true meaining of the instruction coded by 0x90 is 'xchg ax,ax', while 0x91 codes for 'xchg ax,cx'. of course everybody uses the most popular opcode synonym : nop
Shub-nigurrath
March 24th, 2005, 09:16
Well, in x86 architectures the NOP is one of the documented official instructions which the CPU uses and for which there's an opcode. How that is implemented by the cpu is another matter..
this is what the intel instruction reference document tells..
Quote:
NOP—No Operation
Performs no operation. This instruction is a one-byte instruction that takes up space in the
instruction stream but does not affect the machine context, except the EIP register.
The NOP instruction is an alias mnemonic for the XCHG (E)AX, (E)AX instruction.
Flags Affected: None.
Exceptions (All Operating Modes): None.
Opcode Instruction Description: 90 NOP No Operation |
ARM doesn't even have this..
bilbo
March 24th, 2005, 09:58
Shub-nigurrath,
naides is pointing out a (nice) outcome from opcodes decoding.
If you look at the opcode for
XCHG Reg, EAX, the manual says it is implemented with one-byte opcode:
90+Reg,
where Reg (according to one of the initial tables of the manual) can assume the values
Code:
0=EAX, 1=ECX, 2=EDX, 3=EBX, 4=ESP, 5=EBP, 6=ESI, 7=EDI
So
90 (90+0) can be indifferently decoded as
NOP, or also as
XCHG EAX,EAX (in 16 bit mode
XCHG AX,AX).
Best regards, bilbo
Shub-nigurrath
March 24th, 2005, 15:41
humm, interesting, indeed, nice being this the situation so the observation is correct..ARM doesn't have an "official" way to implement the NOP (this is directly "suggested" by intel), you have to think one, given the NOP concept as one operation doing nothing, modifying only the EIP.
..fine.
Powered by vBulletin® Version 4.2.2 Copyright © 2018 vBulletin Solutions, Inc. All rights reserved.