GCC 3.X Ada releases

GCC 3.X Ada releases

[Jerry van Dijk]

A while back I made an experimental release of gnatgcc 3.3 for Windows 
available through www.mingw.org. Meanwhile there have been some further
improvements (bringing the number of AVCS failures down to 47). However
I did not see any bug reports, so I wonder if there is any interest in
newer releases...

gr.
Jerry.

-- 
--  Jerry van Dijk   | email: jvandyk@attglobal.net
--  Leiden, Holland  | web:   users.ncrvnet.nl/gmvdijk

Re: GCC 3.X Ada releases

[Adrian Knoth]

Jerry van Dijk <jvandyk@attglobal.net> wrote:

> I did not see any bug reports, so I wonder if there is any interest in
> newer releases...

Sure there is. It is worse to lack support of something, so someone could
think Ada is not a choice for him and use decreases.


-- 
mail: adi@thur.de  	http://adi.thur.de	PGP: v2-key via keyserver

Er war ein Luftikus - Alle nannten ihn Airwin.

Re: GCC 3.X Ada releases

[Victor Porton]

In article <u1y5k30hg.fsf@attglobal.net>,
	Jerry van Dijk <jvandyk@attglobal.net> writes:
> 
> A while back I made an experimental release of gnatgcc 3.3 for Windows 
> available through www.mingw.org. Meanwhile there have been some further
> improvements (bringing the number of AVCS failures down to 47). However
> I did not see any bug reports, so I wonder if there is any interest in
> newer releases...

About my type of interest (I think not only mine): currently the only reasons
which would cause me to  switch to a newer version of GNAT than 3.14p:

1. Bugs found in old releases.

2. Improved speed of programs and speed of compilation, decreased 
executable size.

3. Extensions in GNAT.* library.

4. Better tools and docs provided with the compiler.

5. New good pragmas etc.

It means that even having the possibility to download a never version  of
GNAT, I will do it only before releasing some Ada software, not in the 
meantime. So I have interest but not now.

Also we need some ASIS programs, where AFAIK gcc 3.x does yet not 
support ASIS.

Re: GCC 3.X Ada releases

[Jeffrey Creem]

I have been using the GNAT version from mingw.org...

Just have not hit any bugs yet. (Code I am working on with this is all
fairly straight forward stuff though
so I am not hitting any edge cases of the compiler yet)...

In any case, I am interested in new releases if for no other reason than to
stay current with
others working on the code on different platforms.


"Jerry van Dijk" <jvandyk@attglobal.net> wrote in message
news:u1y5k30hg.fsf@attglobal.net...
>
> A while back I made an experimental release of gnatgcc 3.3 for Windows
> available through www.mingw.org. Meanwhile there have been some further
> improvements (bringing the number of AVCS failures down to 47). However
> I did not see any bug reports, so I wonder if there is any interest in
> newer releases...
>
> gr.
> Jerry.
>
> --
> --  Jerry van Dijk   | email: jvandyk@attglobal.net
> --  Leiden, Holland  | web:   users.ncrvnet.nl/gmvdijk

Re: GCC 3.X Ada releases

[Florian Weimer]

porton@ex-code.com (Victor Porton) writes:

> 2. Improved speed of programs and speed of compilation, decreased 
> executable size.

Using GCC 3 increases compilation time significantly.

> 5. New good pragmas etc.

Pragma Unreferenced is nice. :-)

Re: GCC 3.X Ada releases

[Preben Randhol]

Florian Weimer wrote:
> Using GCC 3 increases compilation time significantly.

But this is because the gcc-gnat is still in development right?
I mean there isn't a reason for gnat with gcc 3 being always slower than
gcc 2.8 ?

-- 
Preben Randhol ------------------------ http://www.pvv.org/~randhol/ --
�There are three things you can do to a woman. You can love her, suffer
 for her, or turn her into literature.�  - Justine, by Lawrence Durrell

Re: GCC 3.X Ada releases

[chris.danx]

Preben Randhol wrote:

> But this is because the gcc-gnat is still in development right?
> I mean there isn't a reason for gnat with gcc 3 being always slower than
> gcc 2.8 ?


gcc itself is slower that way back then.  That could be having a knock 
on effect on gnat.


-- 
for personal replies change spamoff to chris

Re: GCC 3.X Ada releases

[Preben Randhol]

chris.danx wrote:
> gcc itself is slower that way back then.  That could be having a knock 
> on effect on gnat.

OK. But I guess it is better at optimising?

-- 
Preben Randhol ------------------------ http://www.pvv.org/~randhol/ --
�There are three things you can do to a woman. You can love her, suffer
 for her, or turn her into literature.�  - Justine, by Lawrence Durrell

Re: GCC 3.X Ada releases

[Eric G. Miller]

In <u1y5k30hg.fsf@attglobal.net>, Jerry van Dijk wrote:

> 
> A while back I made an experimental release of gnatgcc 3.3 for Windows 
> available through www.mingw.org. Meanwhile there have been some further
> improvements (bringing the number of AVCS failures down to 47). However
> I did not see any bug reports, so I wonder if there is any interest in
> newer releases...

I have used it a little recently.  Didn't hit any bugs (yet).  So, there's
interest...

Re: GCC 3.X Ada releases

[Simon Wright]

Preben Randhol <randhol+news@pvv.org> writes:

> chris.danx wrote:
> > gcc itself is slower that way back then.  That could be having a knock 
> > on effect on gnat.
> 
> OK. But I guess it is better at optimising?

I got the impression from snooping round the mailing list archives
that the slowdown was related to introducing new support (possibly for
GC within the compiler) rather more then super optimisation.

Re: GCC 3.X Ada releases

[Preben Randhol]

Simon Wright wrote:
> I got the impression from snooping round the mailing list archives
> that the slowdown was related to introducing new support (possibly for
> GC within the compiler) rather more then super optimisation.

I see. But still I expect Gcc 3 to be an improvement over 2.8.1 in the
code generation part :-)

-- 
Preben Randhol ------------------------ http://www.pvv.org/~randhol/ --
�There are three things you can do to a woman. You can love her, suffer
 for her, or turn her into literature.�  - Justine, by Lawrence Durrell