ByteCode vs MachineCode reverse enginering

[Arne Vajhøj]

That site leaves out IBM's and Oracle's, to name two.

True.

But their Java compiler are not interesting either. It is their
JVM's that are different.

For all I know they may even use SUN code for the compiler
(I believe SUN sells commercial licenses for the code).

Arne

 

[Mike Schilling]

True.

But their Java compiler are not interesting either. It is their
JVM's that are different.

For all I know they may even use SUN code for the compiler
(I believe SUN sells commercial licenses for the code)

I don't know why they wouldn't license the Sun compiler; as you point
out, there are no interesting improvements to be made in it..(In the
generation of instructions, anyway. Writing, say,
compile-time-generated hints to their JIT to the constant pool might
be valuable. (Adding compiler-specific attributes is specifically
allowed by
http://java.sun.com/docs/books/jvms/second_edition/html/ClassFile.doc.html#16733,
in case anyone was wondering)).

 

[Mike Schilling]

It means B8 plus the representations of the arguments, which in this
case must be 38.

3B, of course.

 

[Skybuck Flying]

3B, of course.

Well at least you got the 3 correct

Bye,
Skybuck

 

[Roedy Green]

If you play with a machine code disassembler, I think the problems
will come clearer

see http://mindprod.com/jgloss/disassembler.html
--
Roedy Green Canadian Mind Products
http://mindprod.com
"Humanity is conducting an unintended, uncontrolled, globally pervasive experiment
whose ultimate consequences could be second only to global nuclear war."
~ Environment Canada (The Canadian equivalent of the EPA on global warming)

 

[John B. Matthews]

[...]

see http://mindprod.com/jgloss/disassembler.html

I'm not sure what is going on but there are some links on your
website that appear to terminate my browser. The one above is an
example [...]. I'm using Firefox 3.0.3 - any idea what's happening?

[...]

I am unable to reproduce this with Firefox 3.0.4, Safari 3.2.1 or curl
7.16.3.

 

[Lars Enderin]

[...]

see http://mindprod.com/jgloss/disassembler.html

I'm not sure what is going on but there are some links on your
website that appear to terminate my browser. The one above is an
example [...]. I'm using Firefox 3.0.3 - any idea what's happening?

[...]

I am unable to reproduce this with Firefox 3.0.4, Safari 3.2.1 or curl
7.16.3.

Probably it's the applets, which I had to "trust" manually with my Firefox.

 

[Arne Vajhøj]

I don't know why they wouldn't license the Sun compiler; as you point
out, there are no interesting improvements to be made in it..(In the
generation of instructions, anyway. Writing, say,
compile-time-generated hints to their JIT to the constant pool might
be valuable. (Adding compiler-specific attributes is specifically
allowed by
http://java.sun.com/docs/books/jvms/second_edition/html/ClassFile.doc.html#16733,
in case anyone was wondering)).

Considering how commonly used it is to develope and build with SUN Java
and deploy on WAS or WL with other JVM's, then I don't think even that
would be something they would do.

Arne

 

[Mark Thornton]

Hello,

It seems like ByteCode is more easily reversed engineered than MachineCode ?

While it is easier to decompile, I think the implications of this are
quite modest. There two main reasons for not wanting your code decompiled:

1. To prevent bypassing of software protection schemes.
2. To protect valuable algorithms.

In the first case machine code seems to be little impediment. I suspect
that breaking the protection is usually achieved while decoding only a
very small percentage of the complete application.

In the second case, most new algorithms now would be quite large and
difficult to comprehend in the absence of comments and other internal
documentation. If the essence of the algorithm is sufficiently compact
and valuable then machine code will again prove little impediment to the
escape of the secret.

If your secret is really important, then the question of byte code vs
machine code is of little relevance to its protection.


Mark Thornton