Re: Newcomers to

[Simon Clubley <clubley@remove_me.eisner.decus.org-Earth.UFP>]

On 2014-10-10, Natasha Kerensikova <lithiumcat@instinctive.eu> wrote:
> Hello,
>
> On 2014-10-09, Simon Clubley <clubley@remove_me.eisner.decus.org-Earth.UFP> wrote:
>> The problem with FSF GNAT is that once you try to use it with something
>> other than mainstream desktop style operating systems you run into
>> major problems or find outright it simply doesn't work.
>
> What kind of major problems?
>
> I've recently built avr-gnat and it haven't encountered any difficulty.
> Are you implying major issues are awaiting me further in this path?
>

That's because the AVR-Ada project is a third party project and they
have gone to a great deal of effort to support Ada on the AVR.

My comment referred to the kit you download directly from the FSF.
You can directly build C cross and native compilers from that kit for
a wide range of platforms; you can't do that for Ada.

BTW, in case you are interested, Brian Drummond has brought up an Ada
cross compiler for the MSP430. Further information at:

	http://sourceforge.net/p/msp430ada/wiki/Home/

I've also tried using the latest FSF GNAT kit to bring up Ada on VMS
and failed. ACT have their own source tree for the GNAT Pro product
for VMS and they push changes into the FSF tree but it appears for VMS
that either (a) not all the required changes are there or (b) you need
a _very_ specific (and undocumented) combination of build flags.

I don't know which one it is for sure, but I'm tending towards (a).

> I'm also trying to build a cross-compiler for Nintendo DS, it's a bit
> less smooth but I haven't spent much time on it yet. Is it also doomed?
>

I strongly recommend you follow Luke's link; he's spent quite a bit
of effort over the years on this and he's encountered a good range
of problems (there was a time when I was following Luke's work
closely).

The problem is that even to myself (who _wants_ to use Ada) the
language is great but the compiler situation is very flaky especially
when you use a good range of embedded MCUs and non-mainstream
operating systems as I do.

I can only imagine how this looks to a newcomer to Ada especially as
I use C for my embedded projects even when the desire is to use Ada.

In my case, the thing which really forced this issue for me was a few
years ago when I wanted to develop a common bare metal I/O library
plus some other code which worked across _all_ the MCUs I use or might
_want_ to use in the future. (The library has architecture neutral
upper layers and an architecture/MCU specific lower layer.)

I realised the only viable option was C and I realised this was also
true for my embedded projects in general because I couldn't be sure
what future environments I might want to run the code in.

It's experiences like that which make we wonder if we need some kind
of new half-way language between C and Ada which is easy to bring up
on a new target and while not been as type-safe as Ada would be much
better than C.

Simon.

-- 
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world