[ale] assembly

John M. Mills jmills at jmills.gtri.gatech.edu
Mon Oct 5 09:57:31 EDT 1998


On Mon, 5 Oct 1998, Mark Cantrell wrote:

> I don't know if I'm asking the right question so please forgive me.  I
> wanted to study assembly and/or embedding systems.  Does Linux provide a
> good environment for assembly?  Is there some sort of embedded system
> emulator thingy available that I could play around with on my Linux
> machine to learn with.

Mark -

I haven't used these tools, so I may not be on target.  With regard to
assembly language, the default GNU assembler uses a syntax which is not
widely used in the Intel world, but there is an alternative which uses
"Intel" syntax and mnemonics.  It is 'as86' -- I don't find either man
pages nor info on it, so I can only suggest the HOWTOs.  Of course the GNU
tools are highly retargetable, so they are a good base for cross
development.

I don't think that Linux (or any *NIX) is a good native embedded-processor
target, as you must drag a lot of baggage along, defeat the memory
manager, and deal with a kernel of several MBy when you consider ROMing. 
Also embedded applications are often associated with tight timing
requirements: another *NIX weakness.  There is a RT_Linux project (and
some of the associated files can be found on my workstation in: 
 ftp://jmills.gtri.gatech.edu/
in:
 pub/coding/rt_linux
 - not very current, I fear).  Not to say it can't be done, nor that there
are never reasons to undertake it, but you need to have a good mastery of
both embedded apps and your kernel to pull it off, IMHO.

On the other hand, I think Linux is a very promising cross-development
platform.  You would have the important advantages of a strong assemblers,
compilers, debuggers, source management, etc., and would need some
suitable monitor-based debugger which could run on your target boards, of
whatever processor types.  We did this with VME cards under the DEC
debuggers, with tcp/ip between host and target.  I don't know of any
free cross-IDEs for embedded apps, but most of the pieces are there.

I would look in that direction.

  John M. Mills, Senior Research Engineer -- john.mills at gtri.gatech.edu
  Georgia Tech Research Institute, Georgia Tech, Atlanta, GA 30332-0834
         Phone contacts: 404.894.0151 (voice), 404.894.6285 (FAX)
             "Lies, Damned Lies, Statistics, and Simulations"






More information about the Ale mailing list