GCC 3.3.1 Switch Coming Soon On NetBSD

Dan writes "Matthew Green says he is ready to switch sparc, sparc64, i386 & alpha ports to using GCC3.3.1 by default on NetBSD. He's uploaded 4 snapshots (one per port ;-), all cross compiled from i386-netbsd. However, there appears to be work involved with fixing approximately 193 broken packages, as reported by NetBSD's Jan Schaumann."

Re:It's dead, Jim

[bsd_usr]

That's probably because most BSD users have given up on the Slashdot BSD section. The Slashdot BSD section is overrun by trolls and FUD.

BSD is software which can be used freely by anyone. The only way for BSD to die is for people to completely stop using BSD code, but since BSD code is used in so many other software products it really won't ever die.

It's the immaturity of some of these Slahdot readers that have made the Slashdot BSD section go down the drain. But really, you know what's really sad. That these people have nothing esle to do in their poor excuse of a life that they turn to the BSD section to spout out FUD and repeat the "BSD is Dying" troll.

Why do I come here then? Because I too have nothing better to do. Actually, I guess maybe because sometimes through all the trolls and FUD there are some good posts.

Re:It's dead, Jim

[archen]

Given up? I wouldn't say that. Simply put there isn't anything glamorous happening in the world of BSD from a sensationalist point of view. And lets face it, most of the BSD headlines just aren't that interesting. I mean what is there to say about one of the BSDs moving to GCC 3x?

Re:It's dead, Jim

[bsd_usr]

I think you mean one MORE BSD moving to 3.x, but yeah its' true what you say. There's really nothing all that exciting going on. But that can be said about the rest of the Open Source world. Nothing is all that exciting anymore. Not KDE, not GNOME, not Linux, etc. It's just the same old shit cooked up in a different way.

Hell, it's almost safe to say that there isn't much exciting going on in the rest of the computing world. Seems like things are moving kinda slow these days. I guess the only thing that I'm

Re:It's dead, Jim

[R.Caley]

And lets face it, most of the BSD headlines just aren't that interesting.

There is an old BSD curse: ``May you run an interesting operating system''.

Re:It's dead, Jim

[BSD BullShitDefense]

I am fed up with the troll postings.

I am also very pissed that they aren't filtered out.

Lets fight that bullshit!

First I will examine how easy it is to script trollings into the BSD section, perhaps the Slashdot crew does a little bit of defense. But I doubt it somehow.As they don't communicate, I have to find it out myself.

Next thing is to develop a troll scanner to demonstrate how to detect that bullshit. My guess is that those are very easy to detect.

I am not sure what will be the step after th

Re:It's dead, Jim

[bsd_usr]

Have you checked out OSNews. Seems to be alot more open minded inteligent people there that actually post decent comments under BSD related news. It's not perfect, but it's a lot better than Slashdot's BSD Section.

Re:Two hours into the article...

[__past__]

Dude, you could at least read the story title... This is about NetBSD.

I didnt know gcc3.3 was that mature

[mnmn]

Until recently many packages required gcc2.x for compilation simply because 3.x has been relatively untested. Will this bring some kind of instability in NetBSD if say the compilation is successful but with tonnes of warning messages? Are gcc 2.x and 3.x really that different?

Re:I didnt know gcc3.3 was that mature

[__past__]

The biggest annoyance is that newer GCCs treat multiline string literals (i.e. strings with embeded newline characters as opposed to \n) as errors, instead of just issuing a warning. This is actually conforming behaviour, so packages should really be fixed upstream. Other than that, I never had any problems beyond simple warnings, but of course, YMMV.

Given that GCC got better and better in terms of ISO/ANSI C conformance, most problems are probably bugs that just didn't show up yet because they went with a matching GCC bug. So fixing them will only increase conformance, and hence portability to other compilers.

Re:I didnt know gcc3.3 was that mature

[JDizzy]

Yeah, this is the same issue we had in FreeBSD. Lots of little pendantic issues that used to be a warning become errors. This is, as you point out, an issues with interpritaion of the ANSI C standards. Lots of folks try to say their code works, and it doesn't work due to a bug in the new GCC, but that is just water under the bridge when they figure out that the world has changed, and they have to change their code with it.

Re:I didnt know gcc3.3 was that mature

[iamacat]

Well, it depends on weather the code is really violating the standard or it is, as you say, an interpetation problem. If some part of the standard is not clear to an average programmer and could be interpeted in a certain, useful way, the compiler should try to digest the code rather than showing off author's strict interpetation.

Also, I hope older versions of standard are supported and system header files can somehow indicate their own version so that the old programs actually compile.

Re:I didnt know gcc3.3 was that mature

[vesamies]

Inline functions in GCC2 are brainded. GCC3 makes this (and many more things I'm sure) work right. It's a shame they used that old compiler for so long, but I guess stabilizing a toolchain for 30+ (?) platforms is not that easy.

blackbox

[vesamies]

I'm testing GCC3 and the system is working ok. But the failing packages are really annoying. Missing blackbox...

BSD is Rhyming (Re:The *BSD Wailing Song)

[ThreeFarthingStone]

Wow! BSD is the most fun of dying software. Skilled Anonymous Cowards get to make careful calculations about the number of remaining users, and now BSD even receives poems that rhyme!

It seems that the dead or near-dead, like BSD and Mac OS 9, only seem to get more interesting. Mac OS 9 is almost dead, of course, because it was replaced by Mac OS X, which is BSD-based. But the death of BSD seems to be far more interesting anything else. When Microsoft killed Internet Explorer for the Mac, the only result w

It breaks on VAX

[Anonymous Coward]

Yeah, but some of us using VAX systems have discovered that GCC 3.3 DOES NOT WORK on our systems and nobody seems interested in fixing it.
see:
http://mail-index.netbsd.org/port-vax/2003/09/

(remember to take the space out that slashcode puts in)