Re: [Sdcc-user] RE:DS80C400 support

[Nicolas K. <Nic...@da...>]

Bruce,

We have a C library that exposes the ROM functions from the 400 (which 
include the TINI network stack, and the 400 micro OS) to the Keil C 
compiler. I am trying to port it to SDCC, but it will take a little time 
  to develop (bridging between the C parameter passing to the Assembly 
language parameter passing is a bit tough).

 > If you are doing "real time" (and that is often a vague term) control
 > then C and assy may be your best bet, but it will be a LOT more work
 > and the resulting code will be much harder to maintain.

(Throwing in two cents) The 400 rom has a lot of operating system 
constructs, like a scheduler, signals, memory management, and the like. 
I think that will lower the complexity of code and maintenance.

Nick

Bruce Boyes wrote:
> At 08:45 AM 1/23/2003 -0500, ran...@mo... wrote:
> 
>> Nicolas,
>> Thank you for your response!  I am very much interested in porting the 
>> Keil
>> libraries to SDCC.   In fact I've spent the last few days trying to 
>> figure
>> out how to go about this, and I'm stumped at what to do w/the .A51 
>> files -
>> how to port them to link with SDCC.  I'm sure this will go much faster 
>> with
>> you doing it!
>>
>> We (Schneider Electric) are close to making a decision to incorporate 
>> this
>> chip into a new product, a Modbus Serial to Modbus/TCP gateway device, 
>> but
>> the final choice of hardware has not been made.  A big part of the 
>> decision
>> rests on the development environment, ease of use of the tools, etc. 
>> so any
>> progress you make would be greatly appreciated.
> 
> 
> 
> Sounds interesting. I would *strongly* encourage you to do as much of 
> the development as possible in Java. At least try it and see if or where 
> Java just won't do what you need to do. I'd be interested if there is 
> any aspect of the project where Java, with some native methods, won't be 
> good.
> 
> If you are doing "real time" (and that is often a vague term) control 
> then C and assy may be your best bet, but it will be a LOT more work and 
> the resulting code will be much harder to maintain.
> 
> We've coded 80C390 and other 8051s in C and assy for years. We now try 
> to do as much as possible in Java on TINI and our other native-execution 
> Java systems. On the other hand we just finished a complex system and 
> wound up using Java for about 80% of it, and pure assy code for the 
> other 20%. The assy code runs on a small self contained 8051 with ADC 
> and DAC, and sends serial packets to the main controller which is 100% 
> Java and does all the data processing and user interface. The 8051 
> system is very hard realtime, so that's how we sliced it up.
> 
> I'm sure a number of people on this list would like to hear what you 
> learn along the way.
> 
> Regards
> 
> Bruce Boyes
> 
> 


-- 
Nicolas Kral
Software Engineer
Nic...@da...

This supports reflection, which is the 90s way of writing
self-modifying code.
-- John Aycock at IPC7, during his parsing talk