Path: news.eternal-september.org!eternal-september.org!.POSTED!not-for-mail
From: Mikko <mikko.levanto@iki.fi>
Newsgroups: comp.theory
Subject: Re: The execution trace of HHH1(DDD) shows the divergence
Date: Thu, 19 Jun 2025 11:48:13 +0300
Organization: -
Lines: 41
Message-ID: <1030isc$3ph4v$1@dont-email.me>
References: <1021ii4$3327l$6@dont-email.me> <102u15q$31un8$1@dont-email.me> <102umdt$369b2$11@dont-email.me>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 8bit
Injection-Date: Thu, 19 Jun 2025 10:48:13 +0200 (CEST)
Injection-Info: dont-email.me; posting-host="3795b2b31d9ed69cd0c8b5cdd7fda16b";
	logging-data="3982495"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1/a+wl7OPxmzQ6vZnI519RJ"
User-Agent: Unison/2.2
Cancel-Lock: sha1:iO8lJk9CrKJ0Fv/x5ns8V57P6fc=

On 2025-06-18 15:36:29 +0000, olcott said:

> On 6/18/2025 4:33 AM, Mikko wrote:
>> On 2025-06-07 14:32:36 +0000, olcott said:
>> 
>>> The execution trace of HHH1(DDD) shows the divergence
>>> of DDD emulated by HHH from DDD emulated by HHH1.
>> 
>> The divergence means that at least one of them simulates
>> incorrectly. Next should be determined which one.
> 
> *One might incorrectly guess that*

One might, though it would be incorrect only in the sense that
a guess, even a correct one, is not knowledge.

Instead, it is better to start from the meanings of the words
and infer what can be validly inferred:

An execution trace tells what happens during an execution.
It is correct if what it tells is what happens. Two excution
traces are compatible if there is an execution that is
correctly described by both traces (even if that execution
is not of the behaviour that the traces intend to describe).
Divergence means that the traces are not compatible, i.e,
that they do not describe the same behaviour. If one of
them describes the behaviour of DDD then the other one
does not. The meaning of the word "simulation" is that the
trace of a correct simulaton of DDD also descibes the
behaviour of DDD.

> When one tries to find even a single 100% specific
> instruction that is simulated incorrectly, one fails.

If that happens then the traces don't diverge and the only
differences are presentational, e.g. one of the traces
being more complete or more detailed.

-- 
Mikko