Anyone out there using Ada ?

In message , "(see below)" writes

Really? I thought Occam was designed for parallel & distributed processing?

--
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
\/\/\/\/\ Chris Hills  Staffs  England     /\/\/\/\/
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/

The original Apple Lisa and Macintosh used Pascal, and Pascal calling convention and Pascal strings in the ToolBox. 68K assembly language was also used.

Petter

--
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?

It was, but it was very limited (to disjoint communicating processes).

The tasking features of Ada were loosely based on CSP (the ancestor of occam) but go far beyond occam in generality and power, even in Ada 83. Ada 95 and then Ada 2005 added a slew of additional concurrency and real-time capabilities.

--
Bill Findlay
 chez blueyonder.co.uk

I would love to agree, but never having been able to cure myself of writing bugs, who am I not to accept one of life's small realities?

Thinking is certainly the way to go, but belief in perfection? Perfection as a goal, certainly.

Stephen

--
Stephen Pelc, stephenXXX@mpeforth.com
MicroProcessor Engineering Ltd - More Real, Less Time
133 Hill Lane, Southampton SO15 5AF, England
tel: +44 (0)23 8063 1441, fax: +44 (0)23 8033 9691
web: http://www.mpeforth.com - free VFX Forth downloads

Besides the Mac and Lisa (pointed out by Petter), a lot of early DOS- vintage PC software was written in TurboPascal. In that era, most applications carried all of their device drivers around with them, because DOS was mostly used as a program loader. So yes, applications and devices, both.

Modula-3, which is another Pascal derivative, also had explicit support for low-level programming and device-driver writing. Operating systems were written in it.

Cheers,

--
Andrew

I did. I used Pascal for a real-time kernel and various associated hardware drivers that were written in Pascal back around 84-85. Pascal (with just a couple extensions) worked brilliantly for low-level hardware stuff.

--
Grant Edwards                   grante             Yow!  Vote for ME
                                  at               -- I'm well-tapered,
                               visi.com            half-cocked, ill-conceived
                                                   and TAX-DEFERRED!

I've done CS8900 Ethernet device drivers (including the interrupt routine) using Borland Pascal 7. I would use Pascal for everything except that C is the lowest common denominator and is available for just about every chip.

Hi John,

I use Ada on a daily basis at work (not on the MIPS, however). We use it on many of our projects. I also use it on robotics hobby projects at home and do a lot of Windows GUI-based hardware simulators using Ada.

--
Jerry

Yes, heard of both. Maybe you should consider looking into GNAT and publicly documenting any process regarding porting to another piece of hardware, with(out) runtime. This is something I intend to do as it's not my job, just an interest :D but I'm lazy :)

Luke.

At least prior to 1980, I would't even think about writing an ISR in _any_ high level language.

You could never know, what the compiler did behind your back. If for instance the processor did not have multiply instruction, the compiler generated a library call, which usually contained non re-entrant code, which of course is unacceptable.

On VAX/VMS, while a large part of the user mode code (and some executive mode) code was written in Bliss-11 (an intermediate level language of which DEC had full control), most kernel mode and certainly any ISR was written in assembler.

Since 1980, a lot has happened. At least the compiler writer have recognized, that the run time must be re-entrant (which is quite hard in C due to that some string functions having internal storage). Some compiler vendors even have "interrupt" extensions, so they seriously must think about re-entarncy issues in the run time library.

Paul

In 1982 I used a strange Pascal compiler on the Motorola EXORciser targeting the 6809, which had a statically defined 8 level "display register" system to handle nested function definitions.

Such constructs were of course completely useless for handling ISR code.

Paul

I agree (even though I didn't quite agree at the time I learnt Ada and had just been using Occam-2. Ada's advantage is that it was already a complete language when it became available and has been developed ever since. Occam-2, on the other hand, was really just something knocked out by Inmos to support their hardware and basically died when the transputers disappeared.

I was, of course, joking to some extent about the "one true language", but Occam-x could have gone far! You can still get Occam for PC (apparently) here:

It seems that Bliss is much more commonly used language that I have ever expected.

Paul

I am somewhat in the same position as you. While I would probably find doing the port 'fun' there are a few factors about it that would tend to make it a non-trivial port and going without vendor support in this case is somewhat not practical.

Looking into a GNAT paid port for it. Maybe now that PA Semi PowerPC is dead people looking for low power, full temp range, high performance will need to return to the MIPS world (shudder) and vendors will find it more interesting.

On Wed, 30 Apr 2008 18:26:37 -0400, Stephen Pelc wrote: ...

No real disagreement. But I think in general, there are a lot of methods that can be applied to reduce the need to spend a lot of time in the debugger. These range from compiler checks, to use of assertions/DbC where available, coding standards, pair programming, code reviews, test-first "design" and unit tests, frequent integration, to formal methods and even informal reasoning about the code being written. Isn't the common perception that most programmers these days spend most of their time in the debugger?

I have a Cadence Graphics box consisting of an Apple IIe married to an OEM MC68000 coprocessor running an O/S that has drivers written in Pascal ;)

Michael

Is it really true? A debugger should be a tool to help with observation so that you can form a hypothesis from which to generate a yes/no experiment.

From observation of programmers, some can debug, others cannot. The majority of those who can are not great users of debuggers. If you can't observe, you can't debug. Tools that help me observe are the ones I want. When you debug a TCP/IP stack, Ethereal is much more useful than a debugger.

I once debugged a banknote sorting machine (16 bit TMS9995, a long time ago) with a pair of resistor-ladder DACs attached to the upper and lower 8 bits of the address bus and feeding a scope in X-Y mode. That told me far more about my program than any number of debuggers.

Debugging, especially for hard real time systems, has actually become more difficult rather than less over the years. These days, you can manipulate squillions of bits per peekosecond, and there's not a single pin change! And if you attach the usual invasive debugger, your peeksecond becomes a crashosecond.

Stephen

--
Stephen Pelc, stephenXXX@mpeforth.com
MicroProcessor Engineering Ltd - More Real, Less Time
133 Hill Lane, Southampton SO15 5AF, England
tel: +44 (0)23 8063 1441, fax: +44 (0)23 8033 9691
web: http://www.mpeforth.com - free VFX Forth downloads

Perhaps.

I have not used a debugger for over 20 years, and only once or twice in the

10 years before that. Of course, when I did use one, it was for some C code. I have never needed to use a debugger on Pascal, Modula, or Ada programs.

--
Bill Findlay
 chez blueyonder.co.uk

Think you might find it a tad expensive if it's for a personal project.

Not dead, just absorbed by Apple :D

Nothing wrong with MIPS, it's nice n clean...I liked it anyways :P

Luke.