Re: Why RosAsm Breaks on a large number of symbols

From: Betov (betov_at_free.fr)
Date: 07/04/04 

Date: 04 Jul 2004 16:42:41 GMT

"wolfgang kern" <nowhere@nevernet.at> écrivait news:cc9386$12p$1
@newsreader1.utanet.at:

>
> Hi Randy,
>
>| ...But as long as you don't fix the bugs in your assembler,
>
> As I watch this thread for quite a while now,
> I checked what you claim and was just curious "why" Rene resists
> to change a single bit in RosAsm in due of it.

Quite simple, Wolfgang:

Master Self-Esteemed Randall hyde can not live with
the idea that RosAsm is actually the fastest of the
actual Assemblers. This irritates him a lot.

So, he is going to play his same good old bad tricks
as he always did, in the past to "prove" that RosAsm
is slower than... MASM... (???!!!...). (Comparing to,
say, FASM would make Master Pdf a bit less ridicoulous,
but, well... if he likes so... ;)

The evident only way to succeed in such a demential
attempt is to build what he calls, without any sense
of decency, "BenchMarks". That is, tests that are made,
on purpose, weird, by testing tiny parts of the process
and/or by comparing things that can not be compared.

He already did this, in the past (whereas RosAsm was
already clearly much faster than MASM), with some
impressive success, because the guys reading his
bullshits... _do_ believe him. This fact seems to me
incredible, that such an unsignificant ass-hole, having
no idea of what Assembly is, having never done anything
for Assembly, and having nothing to show but undecent
propaganda to sell himself, could be consider someone
who has something to say, but these are real observed
facts.

So, helping master Pdf to understand what he is doing
wrong with his stupid labels List, and why RosAsm will
not compile this, is, indirectely, helping him to build
his next coming durty vile trick.

(Though, i am a bit surprised you did not understood
the "why" when reading RosAsm Source. You have read in
the proper part, but you seem to have "read over" the
point... ;) Ludwig found it very easily...
tssskkk!...
tssskkk!...
tssskkk!... :))

> A short glimpse onto the source (didn't check in deep detail
> as I'm not very familiar with the macros) make me think that
> 'your problem' with this label-list may be not a wrong hashing method
> (the former false positive 'double-symbol' seems to be fixed already),
> it merely looks as the number of 'alias-symbols' cause the hang.
> Your list of labels are compiled to just one label,
> ... with much too many 'me too' names , somehow useless ;) .
> Perhaps you try to insert some code in-between the labels and try
again.

Yes. Evidently. ;) If, for example, you take his famous
Labels List, and that you replace all of the ":" by
": mov eax ebx", you will see that it compiles OK.

Notice also that the given timing, on such a small file
does not mean a thing (under 40 Milliseconds, the times
may vary a lot...).

> Sorry Rene,
> I'm afraid you wanted to keep Randy busy on that for a while ... :)

Yes. I hate you. :)

Not important, i am sure he would find out some way or
some other way, anyway... ;) But,... as long as we can
have fun with this, better is letting him making a fool
of himself as long as possible... :)

Betov.

< http://betov.free.fr/RosAsm.html >

Loading