flat assembler
Message board for the users of flat assembler.
![]() Goto page 1, 2 Next |
Author |
|
LocoDelAssembly 27 Feb 2009, 15:11
Good luck!
BTW, I had the opportunity some months ago to test you hello world example on a friend's smart cell phone when he came to visit ![]() |
|||
![]() |
|
revolution 27 Feb 2009, 15:49
LocoDelAssembly wrote: Good luck! LocoDelAssembly wrote: BTW, I had the opportunity some months ago to test you hello world example on a friend's smart cell phone when he came to visit |
|||
![]() |
|
LocoDelAssembly 27 Feb 2009, 16:05
It greeted the world as expected
![]() |
|||
![]() |
|
vid 27 Feb 2009, 16:07
Congratulations for success so far, and good luck for future. I hope ARM will learn from Intel and release as much as they do.
|
|||
![]() |
|
revolution 28 Feb 2009, 13:41
vid wrote: Congratulations for success so far, and good luck for future. vid wrote: I hope ARM will learn from Intel and release as much as they do. |
|||
![]() |
|
revolution 07 Mar 2009, 03:02
|
|||
![]() |
|
sleepsleep 07 Mar 2009, 06:29
congratulation revolution
![]() ![]() this is what i really looking forward from them. http://www.crunchgear.com/2009/03/02/touch-book-tablet-netbook-with-arm-cpu-10-hour-battery-detachable-screen/ 10 hour battery ![]() i really looking forward !!!, this will possibly change the world. |
|||
![]() |
|
pelaillo 09 Mar 2009, 15:24
Quote:
"... a small step from a man, a big step for the mankind..." Pun aside, it's a very big accomplishment what you have done. Such an acknowledgment clearly indicates the level of quality of fasmarm. Congratulations and thanks for your hard work!! |
|||
![]() |
|
revolution 09 Mar 2009, 15:41
Thanks, but let's not forget the person behind it, Tomasz. It wouldn't be possible without fasm as the base.
|
|||
![]() |
|
Picnic 11 Mar 2009, 21:21
Congratulations revolution, i wish you the best for the future.
|
|||
![]() |
|
revolution 17 Mar 2010, 11:05
The v7 docs (and other stuff) are now publicly released. You still need to sign in to get access but they are free. So far I have never received any unsolicited emails from ARM so using your real email address to sign in shouldn't be a problem. But if you are worried you can always use one of the multitude of throwaway ones around.
You can start here if you don't want to waste time navigating the site menus and whatnot. http://infocenter.arm.com/help/advanced/indexList.jsp It took one year to get to this stage. Another small step. Next stop is completely open access without having to surrender your email address. We shall see. |
|||
![]() |
|
tom tobias 17 Mar 2010, 12:15
revolution wrote: It took one year to get to this stage. Another small step. Next stop is completely open access without having to surrender your email address. We shall see. Thank you. Nice accomplishment. I was able to download a couple of docs and a tutorial, without registering, yet. Umm. I haven't yet read any of the literature, perhaps the answer to my question is found in one of those docs sitting on my hard drive even at this moment... Why is it advantageous to employ an Intel based assembler/compiler in developing an application using RISC architecture, the antithesis of CISC in the Intel world...??? If one sought to employ, for example, the now-dated Motorola/IBM/Apple cpu for some project, would it have made sense to develop such code on an Intel machine??? I guess my question really boils down to why FASM and not RISC-ASM? The very word, FLAT, in Flat Assembler, i.e. FASM, is required because of the incredibly awkward hardware design of the cpu. One is obliged to use "FLAT", because of the arcane insistence by Intel in maintaining the long since obsolete concept of segmented memory--yet it is precisely that awful segmented memory which permeates the core of Intel assembly language. Why would ANYONE seek to impose such a dreadful requirement on a smooth surfaced RISC architecture, for which all memory is flat, i.e. accessible, by definition? That's a mystery to me. Maybe after I read the manual, I will appreciate better your approach, revolution, but having not yet read the docs, it seems to me, something of a hollow accomplishment. I would rather have seen Tomasz' genius directed towards writing a true RISC assembler, designed from the outset to match the internal architecture of that cpu, rather than modifying FASM to represent the realities of RISC. Ah, the bliss of ignorance.... ![]() |
|||
![]() |
|
revolution 17 Mar 2010, 12:24
tom tobias: There is no convenient ARM platform available to do application development. ARM CPUs are generally used for embedded and low power power tasks. Trying to make an ARM system and hook up a monitor, keyboard, HDD, Network etc. could be done (and has been done) but it is not a common setup for many and would have a very limited audience of potential users and uses.
fasmarm is far more useful as a cross-assembler. Develop on one's main (x86) system, test in a VM on one's main system, then transfer to the target and run. I have no plans to make an ARM port of fasmarm. It just doesn't make sense to do that. |
|||
![]() |
|
vid 17 Mar 2010, 12:32
tom: so, FASM design is "flat", RISC architecture memory is flat, where's the problem? Is it just the name of assembler?
|
|||
![]() |
|
tom tobias 17 Mar 2010, 13:06
Thanks vid, thanks revolution.
Let me try again. first, vid, NO. It is certainly NOT the "name" that is aberrant. FASM, FLAT assembler, is juxtaposed to any other INTEL assembler, conceptually, by emphasizing the potential utility of this wonderful tool, in protected mode programming, where, unlike real mode assembly, one has access to the entire range of memory, without segmentation. However, the assembler itself, which autoassembles itself, is obliged to perpetuate all of the conventions of the first iteration of this cpu, i.e. 8086. Every aspect of the traditional CISC architecture is maintained, and available for use with FASM, as it must be, for it is designed to be an Intel cpu assembler. 90% of FASM is utterly useless for any RISC cpu. Suppose the world were different. Imagine that Intel/AMD cpus consituted not 95%, but rather, only 50% of the world cpu market, with ARM cpus comprising the remaining 50%. Would we then employ FASM to create programs running on ARM cpus? I don't think so. revolution wrote: fasmarm is far more useful as a cross-assembler. Develop on one's main (x86) system, test in a VM on one's main system, then transfer to the target and run. Fine. With such an approach, there would be no apple computer. Even FASM itself, has no rationale, if one employs simple economics to drive the equation. We should begin this discussion with the cpu. Is ARM a PROPER cpu for embedded applications? Is it superior to all, or at least MOST other cpus intended for embedded applications? How is it superior? What are its strengths, its competitive advantages, its weaknesses? Why does ARM not enjoy monopoly position in the embedded market, if it is superior to all other low power, low cost cpus of its type, function, and purpose? If this cpu represents the optimum design, the lowest cost/functionality, the lowest cost/operating electrical power, the lowest cost/computational ability, i.e. if this is the best of the lot, THEN, it is unreasonable to NOT develop a proper ARM development system, rather than rely upon tools designed to function with an Apple computer, or a Nokia telephone, or an IBM desktop PC. We had this discussion thirty years ago, at Motorola, regarding the 6809. Some wanted to have a 68000 cross assembler. Some wanted a 6809 development board with power supply and assembler. Some wanted to hop over to the Intel 8086 based IBM PC. I didn't understand the argument then, either!!! ![]() |
|||
![]() |
|
revolution 17 Mar 2010, 13:33
tom tobais: My products use ARM CPUs. That is all. And is the only reason I made an ARM cross-assembler. Using ARM suits the purpose for which they are used. I also have other products with other CPUs (PIC) but the Microchip cross-assembler is good enough that I don't need to make a fasm version for that.
There is no such thing as a "PROPER cpu for embedded applications". It all depends upon the actual application you have in mind. Embedded environments vary greatly from one the another. Each application has its own requirements and thus needs the use of a different CPU to fulfil those requirements. |
|||
![]() |
|
Borsuc 17 Mar 2010, 14:35
tom tobias wrote: If this cpu represents the optimum design, the lowest cost/functionality, the lowest cost/operating electrical power, the lowest cost/computational ability, i.e. if this is the best of the lot, THEN, it is unreasonable to NOT develop a proper ARM development system, rather than rely upon tools designed to function with an Apple computer, or a Nokia telephone, or an IBM desktop PC. ![]() Do you know that most products are developed in factories and used on something much less powered? I'm pretty sure that in the development of AMD CPUs they might use IBM desktop PCs for all sorts of intensive engineering applications. Or to surf the web for information. Or whatever. _________________ Previously known as The_Grey_Beast |
|||
![]() |
|
MazeGen 17 Mar 2010, 15:06
Thanks for the link, revolution. Most of the documents seem to be freely downloadable without registration.
Jazelle v1 Architecture Reference Manual |
|||
![]() |
|
Dex4u 18 Mar 2010, 18:31
A bit late, but congratulation revolution.
|
|||
![]() |
|
Goto page 1, 2 Next < Last Thread | Next Thread > |
Forum Rules:
|
Copyright © 1999-2025, Tomasz Grysztar. Also on GitHub, YouTube.
Website powered by rwasa.