Re: Reverse engineering != piracy (was Re: RosAsm disassembler output vs. IDA Pro)

From: Randall Hyde (randyhyde_at_earthlink.net)
Date: 01/28/04 

Date: Wed, 28 Jan 2004 03:18:07 GMT

"Gerhard W. Gruber" <sparhawk@gmx.at> wrote in message
news:91ed10l20l1svcch8lbmo9nfa7i8g91em0@4ax.com...
> On Tue, 27 Jan 2004 10:15:16 -0000 wrote "Beth"
> <BethStone21@hotmail.NOSPICEDHAM.com> in alt.lang.asm with
> <VYqRb.72$sR3.18@newsfep3-gui.server.ntli.net>
>
> >Indeed; But Rene's "real" assembler doesn't support structures because
> >they are "anti-assembly"...as for constants, they don't currently
> >appear to be "anti-assembly" but I suppose he can _excuse_ this by
> >suggesting hard-coding constants is "specific assembly"...
>
> That doesn't help. Even if he doesn't support structures by itself, he
still
> have to have some notion about it, because a structure requires a certain
> amount of memory. If the code, which is disassembled, will only manipulate
> some fields of that structure, then how does the person, disassembling the
> code, know how large the memoryblock should be? I guess code which is
> potentially to be disassmbled will be some stuff which is hard to write
> yourself like decoding an audio/videostream or similar things. And this
code
> will get som buffer which may have to have a certain size, which the guy
will
> have no knowledge about, unless he also disassembles the caller. That's
where
> I say the usefullness will stop, because, even if someone is determined
enough
> to do some disassembling, there is only so much time you can sepnd on it.
> Considering how well wine does, and how many people are involved, you can
> easily determine how well a lone person will do wich HAS to disassemble
some
> code just to get some functionality. Either he will switch to another IDE
> which makes it easier, or he will look for some open code he can adapt.

It gets even worse. One of the main reasons for disassembling code is so you
can *modify* it afterwards. Guess what happens to all those "sizeof(
somestruct)"
expansions in the disassembled code? They won't get adjusted to deal with
the
new size of the structure if you decide to change the structure's size
(assuming
you code even extract that information in the first place).

Rene continues to believe that an automatic disassembler is a perfectly
reasonable thing and that he will succeed, even though the concept is
a proven mathematical impossibility (it violates the "halting theorem").
Oh well, it's his time he's wasting. He's been told the real deal by lots of
different people, if he wants to keep tilting at windmills, that's his
business.
Cheers,
Randy Hyde

Loading