How to keep the C code produced by GNAT ?

How to keep the C code produced by GNAT ?

[Gautier]

Thank you for answering such a simple question...

Re: How to keep the C code produced by GNAT ?

[Dale Pontius]

In article <1996Dec6.153517.5380@news>,
        Gautier.DeMontmollin@maths.unine.ch (Gautier) writes:
>Thank you for answering such a simple question...

I don't believe C code ever gets generated by GNAT. I believe
the Ada front-end feeds abstract data structures to the gcc
back-end directly. No C ever exists.

Dale Pontius
(NOT speaking for IBM)

Re: How to keep the C code produced by GNAT ?

[Adam Tyner]

In article <58a36c$13ra@mdnews.btv.ibm.com>, pontius@btv.ibm.com says...
>
>I don't believe C code ever gets generated by GNAT. I believe
>the Ada front-end feeds abstract data structures to the gcc
>back-end directly. No C ever exists.

What I believe he's thinking of is there is a file with a .C extension that's
produced when GNAT compiles.  I've edited one of these .C files that was left 
when GNAT crashed once, and Dale's right -- it's not C, despite the extension.

-Adam Tyner

-- 
/----=========================================----\
     http://www.awod.com/gallery/rwav/ctyner/
The home of He-Man, "Weird Al", Yoo-Hoo, and more!
ctyner@awod.com           O-         MiSTie #67,326

Re: How to keep the C code produced by GNAT ?

[Robert Dewar]

Adam Tyner said

"What I believe he's thinking of is there is a file with a .C extension that's
produced when GNAT compiles.  I've edited one of these .C files that was left
when GNAT crashed once, and Dale's right -- it's not C, despite the extension."

That's odd, no ijntermediate files with an exception of .C are generated
to my knowledge, other than the binder generated main program file.

Re: How to keep the C code produced by GNAT ?

[David C. Hoos, Sr.]

Hi all,
The program gnatbl is provided for binding and linking using the RTL in the
most simple situations.  Gnatbl calls  gnatbind which creates a C file
(b_hello.c, for example) containing calls to all of the elaboration
routines; then it compiles this file; then it invokes gcc to do the
linking.  If the switch -g is used, debugging information is included in
the objects, and the b_<main-unit-name>.c file is kept for use by symbolic
debuggers.  To my knowledge, this is the only .c file produced by any of
the gnat programs.

I hope this makes the matter more clear.
 
David C. Hoos, Sr.,
http://www.dbhwww.com
http://www.ada95.com

Adam Tyner <ctyner@awod.com> wrote in article
<58fbmg$guc@hubcap.clemson.edu>...
> In article <58a36c$13ra@mdnews.btv.ibm.com>, pontius@btv.ibm.com says...
> >
> >I don't believe C code ever gets generated by GNAT. I believe
> >the Ada front-end feeds abstract data structures to the gcc
> >back-end directly. No C ever exists.
> 
> What I believe he's thinking of is there is a file with a .C extension
that's
> produced when GNAT compiles.  I've edited one of these .C files that was
left 
> when GNAT crashed once, and Dale's right -- it's not C, despite the
extension.
> 
> -Adam Tyner
> 
> -- 
> /----=========================================----\
>      http://www.awod.com/gallery/rwav/ctyner/
> The home of He-Man, "Weird Al", Yoo-Hoo, and more!
> ctyner@awod.com           O-         MiSTie #67,326
> 
>