* GCC 3.4 Ada compiler quality?
@ 2004-04-21 18:15 Brian Catlin
2004-04-21 18:52 ` Adrian Knoth
2004-04-22 11:57 ` Georg Bauhaus
0 siblings, 2 replies; 4+ messages in thread
From: Brian Catlin @ 2004-04-21 18:15 UTC (permalink / raw)
I know it was just released, but does anyone have a good feel for the quality of
the new GCC 3.4 Ada compiler? Is it worth upgrading from 3.15P, or should I
wait until 3.5?
-Brian
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: GCC 3.4 Ada compiler quality?
2004-04-21 18:15 GCC 3.4 Ada compiler quality? Brian Catlin
@ 2004-04-21 18:52 ` Adrian Knoth
2004-04-22 11:57 ` Georg Bauhaus
1 sibling, 0 replies; 4+ messages in thread
From: Adrian Knoth @ 2004-04-21 18:52 UTC (permalink / raw)
Brian Catlin <BrianC@sannas.org.bad> wrote:
> the new GCC 3.4 Ada compiler? Is it worth upgrading from 3.15P, or should I
> wait until 3.5?
I'd been using the 3.4-cvs for a while. Six months ago it was fine and did
the job. I think is is worth upgrading, but I don't know why ;) IIRC they
changed a lot in procedure-calls. Instead of pushing the parameters to
the stack and popping them afterwards they're calculating the variable's
address and do a normal subroutine-call afterwards. But consider this
as highly iirc, I don't even know if it's true at all ;)
Someone contributed graph-coloring-optimization-code, I guess it's for
automatically deciding which variables to use as register-vars (faster
than RAM-vars). But I'm not sure ;)
Besides this a bugtracking-system exists and I'd guess the Ada-component
itself was improved due to userfeedback. Dunno.
--
mail: adi@thur.de http://adi.thur.de PGP: v2-key via keyserver
Q: Was passiert, wenn eine deutsche Blondine nach �sterreich auswandert?
A: In beiden L�ndern steigt der durchschnittliche IQ. (aus de.talk.jokes)
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: GCC 3.4 Ada compiler quality?
2004-04-21 18:15 GCC 3.4 Ada compiler quality? Brian Catlin
2004-04-21 18:52 ` Adrian Knoth
@ 2004-04-22 11:57 ` Georg Bauhaus
2004-04-22 12:44 ` Duncan Sands
1 sibling, 1 reply; 4+ messages in thread
From: Georg Bauhaus @ 2004-04-22 11:57 UTC (permalink / raw)
Brian Catlin <BrianC@sannas.org.bad> wrote:
: I know it was just released, but does anyone have a good feel for the quality of
: the new GCC 3.4 Ada compiler? Is it worth upgrading from 3.15P, or should I
: wait until 3.5?
It has solved some problems for me having to do with private
children, and with generic formals in certain circumstances.
I find -gnatwa quite useful
The project file support is better (though this might affect
portabiliy).
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: GCC 3.4 Ada compiler quality?
2004-04-22 11:57 ` Georg Bauhaus
@ 2004-04-22 12:44 ` Duncan Sands
0 siblings, 0 replies; 4+ messages in thread
From: Duncan Sands @ 2004-04-22 12:44 UTC (permalink / raw)
To: Georg Bauhaus, comp.lang.ada
On Thursday 22 April 2004 13:57, Georg Bauhaus wrote:
> Brian Catlin <BrianC@sannas.org.bad> wrote:
> : I know it was just released, but does anyone have a good feel for the
> : quality of the new GCC 3.4 Ada compiler? Is it worth upgrading from
> : 3.15P, or should I wait until 3.5?
>
> It has solved some problems for me having to do with private
> children, and with generic formals in certain circumstances.
>
> I find -gnatwa quite useful
>
> The project file support is better (though this might affect
> portabiliy).
I have had problems with the code generated when using -gnatn,
and sometimes when just using -O2 (mysterious segmentation
faults, exception handling failing etc). I haven't noticed any regressions
when compiling without optimisation, compared to gnat 3.15p.
All the best,
Duncan.
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2004-04-22 12:44 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2004-04-21 18:15 GCC 3.4 Ada compiler quality? Brian Catlin
2004-04-21 18:52 ` Adrian Knoth
2004-04-22 11:57 ` Georg Bauhaus
2004-04-22 12:44 ` Duncan Sands
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox