From Newsgroup: comp.lang.c

On 8/9/2025 5:00 PM, Richard Heathfield wrote:

On 09/08/2025 22:38, olcott wrote:

On 8/9/2025 4:26 PM, Richard Heathfield wrote:

On 09/08/2025 22:22, olcott wrote:

On 8/9/2025 4:12 PM, Richard Heathfield wrote:

On 09/08/2025 21:46, olcott wrote:

On 8/9/2025 3:41 PM, Richard Heathfield wrote:

<snip>

You get the wrong numbers out. It don't get much more flawed than >>>>>>> that.

At this point you are essentially saying that
the emulation is flawed because everyone knows
that "push ebp" really means "jmp 00002155".

No, I'm saying it's flawed because everyone knows that 0 != 1.

Those are your only two possible results: it stops, or it
doesn't. If you get the wrong one, your emulation is broken.

Ah so you are dishonest. That is what I expected.

WHAT?

_DD()
[00002162] 55             push ebp
[00002163] 8bec           mov ebp,esp
[00002165] 51             push ecx
[00002166] 6862210000     push 00002162 // push DD
[0000216b] e862f4ffff     call 000015d2 // call HHH
[00002170] 83c404         add esp,+04
[00002173] 8945fc         mov [ebp-04],eax
[00002176] 837dfc00       cmp dword [ebp-04],+00
[0000217a] 7402           jz 0000217e
[0000217c] ebfe           jmp 0000217c
[0000217e] 8b45fc         mov eax,[ebp-04]
[00002181] 8be5           mov esp,ebp
[00002183] 5d             pop ebp
[00002184] c3             ret
Size in bytes:(0035) [00002184]

You have to go through the above code line-by-line
knowing that each time HHH is called it creates a
separate process context to emulate an instance of
DD and then emulate an instance of itself emulating
DD when DD calls HHH(DD).

Within this you must show exactly how the original
emulated DD reaches past its own machine address of
[0000216b].

Why?

Haven't you already done it?

You are the one that claimed:

On 8/9/2025 2:30 PM, Richard Heathfield wrote:

you will have proved that emulation is a flawed technique,

So it is up to you to point out the exact flaw.

Oh, I see.

The exact flaw is that you get a different result from the direct
execution, as you claimed: "the behavior of the directly executed
machine different than the behavior of the correctly emulated machine description."

Yes so within the assumption that they must be the same
I would be incorrect.

I proved that the assumption is false on the basis of
the ultimate measure of correct simulation (the semantics
of the x86 language) and you failed to find any error.

If "push ebp" really means "jmp 00002184" then this
would prove that I am wrong.
--
Copyright 2025 Olcott "Talent hits a target no one else can hit; Genius
hits a target no one else can see." Arthur Schopenhauer
--- Synchronet 3.21a-Linux NewsLink 1.2