home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Reverse Code Engineering RCE CD +sandman 2000
/
ReverseCodeEngineeringRceCdsandman2000.iso
/
RCE
/
LiuTaoTao
/
crhwinfo.txt
< prev
next >
Wrap
Text File
|
2000-05-25
|
2KB
|
54 lines
Software name:
HWiNFO Version 4.2.2 Oct-20-1997
The file 'hwinfo.exe' begin with:
V V V V
0000: 4D 5A 50 01 65 01 00 00 02 00 F3 4B FF FF 00 00 MZP.e......K....
0010: 30 00 00 00 40 00 00 00 1C 00 48 57 69 4E 46 4F 0...@.....HWiNFO
This means it needs
150h*200h + 4bf30h = 75f30h
bytes memory to run ! But we only have 0-9ffff, DOS will use some,
TR will use 120K, How can I trace it in TR ?
Does it really need so much memory ? Lets have a test:
(1) Change 'hwinfo.exe' offset 0ah from 'f3 4b' to 'f3 0b'.
--------------------------------------------------------
Run new hwinfo.exe, OK!
(2) Now lets unpack it:
TR hwinfo.exe ;modified
getknl
MKEXE
This will create file MEM.EXE. It runs OK!
(3) File mem.exe still has a shell. I have tried to unpack it:
copy mem.exe mem1.exe
TR mem1.exe ;please pay attention to SS.
;if SS>a000, you should optimize your memory first
exe1
reload
goreg cs ax=0 ;wait a moment
wexe1 ;wait
exe2
reload
goreg cs ax=0
wexe2
q
MKEXE
This make new MEM.exe, but it doesnt work !
I donot know why the second mem.exe doesnt work, but I think it must
concern to its large size. I have tried UPC(v1.06.3) and TEU(v1.73e),
they all find the file is compiled by Borland C v3, but they all can
not get a usable new EXE file.
TR can successfully unpack the first shell of hwinfo. And TR can unpack
the second shell, but still has some problem to solve. This is what I
can do today.