Prev: [blog] The Henn and Egg problem
Next: Call for contributions to SIGAda 2008, Portland Oct. 26-30
From: Alex R. Mosteo on 22 Jan 2008 16:45
Pascal Obry wrote:

> tmoran(a)acm.org a écrit :
>> No. Does its tasking work under Vista?
>
> Yes, GNAT is working fine on Vista.
>
> Pascal.

I though there were problems with the stack execution protection? I.e. some
constructions would cause programs to be killed (local subprograms?
trampolines?).

That is, in the pre-last version before being fixed.
From: Pascal Obry on 22 Jan 2008 16:54
Alex R. Mosteo a écrit :
> I though there were problems with the stack execution protection? I.e. some
> constructions would cause programs to be killed (local subprograms?
> trampolines?).
>
> That is, in the pre-last version before being fixed.

Sorry I wasn't clear. There were problems indeed because of the DEP. But
this has been fixed.

Pascal.

--

--|------------------------------------------------------
--| Pascal Obry Team-Ada Member
--| 45, rue Gabriel Peri - 78114 Magny Les Hameaux FRANCE
--|------------------------------------------------------
--| http://www.obry.net
--| "The best way to travel is by means of imagination"
--|
--| gpg --keyserver wwwkeys.pgp.net --recv-key C1082595
From: Phaedrus on 22 Jan 2008 18:30
> This is a very common (and unfortunate) line of thinking: "I prefer bugs
> that I know rather than (fewer) bugs that I don't know".

Nothing unfortunate about it. There are multiple paths to success, and
sometimes the path you know leads you through the mud. Nasty, yes, but far
better (From a "staying alive" point of view) than the path you don't know
which could lead you along a nasty cliff, and possibly over! Try rephrasing
this to "When they don't keep me from delivering a product on time, I prefer
the bugs I know to the bugs that might keep me from delivering and getting
paid". Now, see which side of the argument makes more sense. If the bugs
that I know keep me from delivering, or add significant cost/time to the
development project, then I may take the leap. Or, if I can spare the
manpower to carefully try the new stuff under controlled conditions, then
sure. For example, Vista is the latest and greatest piece of Windows
software. Want to bet your development schedule on it's performance?

> Why did I climb on my soapbox? Because I see constantly people asserting
> the risk of making a change, without asserting the risk of *not* making
> the change.
>
> Note that most of the free software community takes the opposite view
> (always use the latest version) and it does not work so bad...

Professional software development is all about managing risk. For hobbyists
(and some academics) it's fine to assume the unknown risk that accompanies
the "latest and greatest" release, but keep in mind that the rest of us have
to keep our customers satisfied by understanding the cost/benefit ratio of
these kinds of decisions.

Brian

"Jean-Pierre Rosen" <rosen(a)adalog.fr> wrote in message
news:pf64nf.bso.ln(a)hunter.axlog.fr...
> tmoran(a)acm.org a �crit :
>> These are old programs that (except for a new program to test for this
>> problem) were compiled some time ago with 3.15p I hate to shake the
>> dust off the source code of working programs and recompile new versions
>> hoping that the old bug will be fixed and no new ones will appear -
>> though I realize I may have to do just that.
> This is a very common (and unfortunate) line of thinking: "I prefer bugs
> that I know rather than (fewer) bugs that I don't know".
>
> Although there is some sense in it, this is just pushing too far. Be
> assured that *lots* of bugs have been fixed since 3.15p, and if new ones
> have been introduced, they are certainly in very complicated and unlikely
> cases, since the latest versions of Gnat are routinely used for demanding
> and critical software - or you can have bugs in the recent 2005 features,
> but you are certainly not using them.
>
> Why did I climb on my soapbox? Because I see constantly people asserting
> the risk of making a change, without asserting the risk of *not* making
> the change.
>
> Note that most of the free software community takes the opposite view
> (always use the latest version) and it does not work so bad...
> --
> ---------------------------------------------------------
> J-P. Rosen (rosen(a)adalog.fr)
> Visit Adalog's web site at http://www.adalog.fr


From: tmoran on 22 Jan 2008 23:30
>> This is a very common (and unfortunate) line of thinking: "I prefer bugs
>> that I know rather than (fewer) bugs that I don't know".
>
>Nothing unfortunate about it. There are multiple paths to success, and
A bug that is known and successfully worked-around isn't a bug, it's
a feature. Probably not a nice feature, but not a nasty surprise waiting
to bite you.
Personally, I usually don't enjoy time spent finding the bugs in
other people's (compilers or OSes etc) code. I usually have little idea
where the problem might be and there's a lot of grunt work trying to
cut down to a small program that demonstrates the problem (see the
recent thread on "Ada obfuscation"). And when there is a simple
example, learning what to avoid may be its only payback, while those
responsible take months, or even years, to release a corrected version.
My understanding is that the current Gnat is not a 3.15p with
improvements, but is rather a substantially new thing, with new,
unknown, bugs. Is that a mis-understanding?
From: Ludovic Brenta on 23 Jan 2008 03:32
tmo...(a)acm.org wrote:
> My understanding is that the current Gnat is not a 3.15p with
> improvements, but is rather a substantially new thing, with new,
> unknown, bugs. Is that a mis-understanding?

IMHO your understanding is correct up to a point, but not for the
current version of GCC (4.2.2) or the next version.

When AdaCore merged GNAT into the GCC sources, a long period of
instability ensued for the reasons you outline. I never considered
versions 3.1, 3.2 or 3.3 production-worthy and this is where I agree
with your understanding. However, 3.4 was OK. With version 4.0, a
major change to the back-end ("tree-SSA") introduced some new bugs in
the interface with the Ada front-end, but 4.1 was OK again and the Ada
part of GCC has steadily improved since (i.e. no or very few
regressions). I have a very good feeling about the upcoming GCC 4.3
because AdaCore and Sam Tardieu have recently fixed several dozen old
bugs, most of which have been present since GNAT 3.15p or before. I
anticipate that GCC 4.3 will be the best version of GCC ever as
regards Ada.

Debian reflects this opinion of mine: in 3.1 "Sarge" the Ada compiler
was GNAT 3.15p, in 4.0 "Etch" it is GCC 4.1, and in the upcoming
"Lenny" it will probably be 4.3.

Of course, none of the above applies to GNAT Pro or GNAT GPL Edition
because AdaCore are careful not to introduce regressions when
switching to a new GCC back-end, and do so only infrequently:

GNAT Pro 3.x: GCC 2.8.1
GNAT Pro 5.x and GNAT GPL 2005 Edition: GCC 3.4
GNAT Pro 6.x and GNAT GPL 2006 and 2007 Edition: GCC 4.1

--
Ludovic Brenta.
First  |  Prev  |  Next  |  Last
Pages: 1 2 3 4
Prev: [blog] The Henn and Egg problem
Next: Call for contributions to SIGAda 2008, Portland Oct. 26-30