Re: Creating a dis-assembler on my own - shucks !
From: Matt Taylor (para_at_tampabay.rr.com)
Date: 01/18/04
- Next message: Phil Carmody: "Re: Non Intel & AMD Arch"
- Previous message: Matt Taylor: "Re: Creating a dis-assembler on my own - shucks !"
- In reply to: Bx. C: "Re: Creating a dis-assembler on my own - shucks !"
- Next in thread: Matt Taylor: "Re: Creating a dis-assembler on my own - shucks !"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Sun, 18 Jan 2004 05:41:08 +0000 (UTC)
"Bx. C" <null@the.void> wrote in message
news:Q9lOb.4207$D%.1175@bignews1.bellsouth.net...
<snip>
> true, a far call or jump to a location specified in a 16-bit reg is
totally
> illogical... but evidently, they took the easy way out... personally, i'd
do
> like you, except, instead of dumping a "db" line for each individual byte,
> i'd go until i hit the next valid instruction, and dump a single line (if
> possible, or groups of 16 per line) for the whole chunk of invalid
> instructions... it's bad enough that doing a disassembly on what's
supposed
> to be standard alphabet text strings yields a ton of single and double
byte
> instructions... (push xxx / pop xxx / dec xxx / inc xxx / jxx xxxx,
etc)...
> i'm trying to think up the best way of identifying long text strings and
> intelligently display them as data instead of instructions, within a
> disassembly...
The easiest answer is to call it a user error. There really is no "best"
way. IDA does a good job of avoiding data by analyzing the code to determine
where the program might execute and assuming all else is data. Unfortunately
this analysis is not perfect, and there are a number of somewhat uncommon
cases which break it. Also, the first thing anyone tries when obfuscating
code is to invoke those degenerate cases which make code & data look the
same. My favorite trick is to insert a conditional branch, preferably
data-driven, that is never taken and points at obfuscated data.
-Matt
- Next message: Phil Carmody: "Re: Non Intel & AMD Arch"
- Previous message: Matt Taylor: "Re: Creating a dis-assembler on my own - shucks !"
- In reply to: Bx. C: "Re: Creating a dis-assembler on my own - shucks !"
- Next in thread: Matt Taylor: "Re: Creating a dis-assembler on my own - shucks !"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|