Re: RosAsm Team is Still Making Excuses


wolfgang kern wrote:

>
> If you like to have a consecutive set of bytes, then just
> declare them within a line.
> [MyBytes: B$ 0 1 2 3 4 5]
> BU_ASM_Help says: 'a new line forces DW-alignment'.
>
> Hard to understand for the 'not that smart' ?

???
Hard to believe, more like. Are you suggesting that

[MyBytes: B$ 0 1 2 3 4 5
6 7]

actually consumes 10 bytes of memory?

Perhaps you quoted it out of context, but that's what the statement you
just made implies.

>
> | It's pretty clear that the RosAsm crowd thinks they are an elite
bunch.
> | Every time a perceived problem is posted to this newsgroup the
response
> | seems to be
>
> | 1. Well, you just don't understand how to use RosAsm.
> | 2. Your IQ isn't high enough to understand what's going on.
> | 3. That could be fixed by one instruction, but we're not going to
do it
> | because we know how to work around the problem and you don't.
> | 4. etc., etc., etc.
>
> There are that many source-code examples of any style,
> that everyone who's really interested in, can find out
> how it works without too many questions.
> Of course a minimum of logical thinking is required.

Forgive me. I forgot:

5. Of course a minimum of logical thinking is required.


> | The interesting part is what this posts *don't* include. Namely,
things
> | like:
> | 1. Here's how you do this in RosAsm
> | 2. Here's an explanation that someone who only has an IQ of 80 can
> | understand.
> | 3. Here's the corrected version of the software (or the work around
> | that we all use).
>
> If you would be seriously interested in RosAsm, then you wont come
> up with that sort of questions.

You're right that I don't have a serious interest in RosAsm. Something
the world at large shares with me, I might point out.

But I think that it's a quite serious question to ask why RosAsm insist
on aligning byte variables on dword boundaries when there is absolutely
no need to do so.


> Ask the (your?) Governor how to succeed with a low IQ :)

I believe that would be the (point #2) response listed above.

Just keep in mind that Arnie is a heck of a lot more successful that
RosAsm will ever be :-)

>
> | This leads the casual reader to assume:
> |
> | 1. The RosAsm team doesn't understand the problem either.
> | 2. The RosAsm team can't figure out this stuff, either.
> | 3. The RosAsm team hasn't got a clue how to fix the problem.
>
> The 'casual' reader is warned on top: "work in progress".

Uh, "in progress"?
How about the quote from Rene: "I'll never fix this problem?"


>
> | So, if you've got some amazingly good information about RosAsm's
> | behaviour that contradicts what the documentation says, feel free
to
> | share it with us.
>
> Yes, the documentation is somehow behind the actual status
> (as usual for projects in progress).
>
> I didn't find any wrong statements in BU_ASM, Ok, some parts are
> missing and a few had to be read twice due a bit French influence.

All documentation for *every* assembler out there is a little behind.
RosAsm is no different on this matter. The state of the documentation
isn't the issue. What I'm saying is that if the documentation says one
thing about the byte alignment issue *AND THIS IS INCORRECT*, feel free
to correct me on that. AFAICT (based on Rene's responses), RosAsm
*still* aligns bytes in a declaration section of their own on a dword
boundary when there is no need for this. He even claims that it would
only take a few instructions to solve this problem. So that kind of
tells me that the documentation *is* in sync with reality on this
point.

And their is no good technical reason for this in RosAsm. It wastes
space (thereby wasting cache space, a precious commodity) and can cause
a program to run a little slower (because of cache misses caused
because the program uses more cache space). As it would cost *so
little* to fix the problem (by Rene's own words), it doesn't make sense
*not* to correct this problem.

Cheers,
Randy Hyde

.

Relevant Pages

  • Re: Linux, X, ld, gcc, linking, shared libraries and stuff
    ... > This is a real problem with SoftWare (not only RosAsm), ... > some quantity of Functionalities, ... the harder it is to "keep up" with the documentation... ... >> powerful macro system (you know, perhaps you don't feel it's better ...
    (alt.lang.asm)
  • Re: RosAsm Team is Still Making Excuses
    ... RosAsm is no different on this matter. ... What I'm saying is that if the documentation says one thing about the byte alignment issue *AND THIS IS INCORRECT*, feel free to correct me on that. ... RosAsm *still* aligns bytes in a declaration section of their own on a dword boundary when there is no need for this. ... because the program uses more cache space). ...
    (alt.lang.asm)
  • Re: Is MASM32 an evil Microsoft plot?
    ... Herbert Kleebauer écrivait ... > and in a last step wrote the Windows assembler version. ... this "Documentation"? ... Now, if _you_ cannot wait, you can run RosAsm, load ...
    (alt.lang.asm)
  • RosAsm Team is Still Making Excuses
    ... >> These two declarations actually wind up consuming eight bytes in ... >> value that RosAsm uses for alignment is incorrect. ... The RosAsm documentation isn't exactly the best ...
    (alt.lang.asm)
  • Re: Structures in Assembly Language
    ... the RosASM audience, and the integrated nature does make for a good ... Pascal was the first language I learned, and I had no problem reading ... easily declare & initialize nested structures in RosASM? ... and it certainly "associates a data type with the name". ...
    (alt.lang.asm)