Philippe Moutarlier wrote:
>
> Michael,
Philippe,
;)
> There are TONS of messages on the net (some from Alan Cox which seems to
> know what he is talking about) about the fact that the gcc 2.96 is NOT a
> release,
True.
I won't repeat Rick and Hal on the kgcc front, but I believe that your
problems have their origin somewhere other than Red Hat's little gcc
compromise, which is why I asked for more detail; the sources I've been
using on RH7 (both Red Hat's tree, once I did "make mrproper", and
pristine tarballs from kernel.org) have compiled flawlessly. The fact
that you're having all these problems leads me to suspect that somehow,
possibly in the process of trying to fix whatever was first causing your
"SMP" errors, you broke something else on your system.
It is my experience that a Red Hat 7 system, whether installed as-is
from the CD release or patched to errata levels as of about three days
ago (can't necessarily swear anything about patch states in between ;) )
is capable of compiling a kernel for itself. I am far too ignorant to
fix a broken compiler myself, even by accident; and if it didn't work, I
promise that this list as well as several others would be screaming
about it. It hasn't been, beyond a bit of a flurry when the news about
7 first came out.
I've only seen one opinion from Alan Cox on this issue; I forget where I
read it, but I seem to recall that he endorsed Red Hat's decision as
innovative and reasonable under the circumstances. Bear in mind that
Alan is a Red Hat employee, but I would hope that his opinions would
inform their decisions, rather than the other way around.
> The gcc web site has a special page about this and is warning that some
> distributions
> are distributing this compiler which is not, and will never produce code
> compatible with either 2.95 or 3.0 .
As I understand those reports, most of the issues surround object code
which was compiled from C++ source being unable to link against object
files from other versions of the compiler. That does not mean it's a
broken compiler, just that we must be cautious as we use it and in some
cases use a different one. Computing is full of compromises, and there
often aren't neat, everybody-wins solutions to problems, particularly
when large projects (like compilers) are undergoing major changes (as
gcc has been over the past several months and seems likely to continue
to do).
> They even claim that such release does
> NOT exist from their point of view.
Red Hat folks have effectively agreed with this, I think.
> Why was the glibc distributed in 6.2 bugged ?
> Why is the glibc distributed in 7.0 bugged ?
According to my informed layman's understanding, because it's huge.
Every version of libc that I've ever dealt with for any period of time,
free or not, has had bugs. Any project the size of libc is going to
have bugs, some of them severe. Even OpenBSD hits a glitch now and
then, and you can't get much more picky than those guys.
Testing can only solve a finite number of problems. Don't give up on
Red Hat because they release buggy software; everybody does that. Give
up on them if the bugs go unsolved.
I don't care whether you use Red Hat or not; but I would hate to see
anybody run off in search of The Perfect Distribution, because it does
not exist, not yet anyway, and in this particular case I remain
convinced that your problems are being caused by something other than
Red Hat's choice to include an odd version of gcc.
> It might happen you did not upgrade gcc.
Installed RH7 on a clean partition. Worked fine.
> Now, it is simple: I deinstall their compiler, go and find another STABLE
> one, recompile my kernel which was not properly installed in the first place
> (why ?)
Don't know; mine was, on four machines so far, one of them ancient, two
less than six months old.
> and I lost a full day "upgrading".
Count your blessings. I've lost many days to upgrades under Windows,
SCO and Solaris; have had much better luck with Red Hat, and _believe_
me, I'm far from an expert. I've had my moments of frustration with Red
Hat and gone looking at other distributions, but I always ended up
coming back for one reason or another.
Just trying to keep things in perspective. If you have better luck with
a different provider, by all means switch; but expect to lose a day now
and then no matter what, until computer science is a lot more mature
than it is today.
> -----Original Message-----
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of Michael R. Jinks
> Sent: Tuesday, December 05, 2000 3:43 PM
> To: [EMAIL PROTECTED]
> Subject: Re: 7.0 "upgrade" pbs ... more and WARNING
>
> Philippe Moutarlier wrote:
> >
> > I respond to myself as it seems that it is a well known pb with RH 7.0 .
> > the shipped version is bugged !!!!!
>
> What makes you say this? Be specific. I've had no trouble using RH7, and
> I've
> now compiled the kernel several times, so I'm inclined to suspect that
> something
> is wrong with your machine, not with the distribution itself. If this is a
> well-known issue, there should be a fix available.
>
> (Note: my RH7 machines have been patched with the Red Hat errata (available
> via
> ftp if you're still having problems with up2date), so well-known bugs
> present on
> the CD may have been fixed for me, and could be for you if you apply the
> updates.)
>
> --
> Michael Jinks, IB // Technical Entity // Saecos Corporation
> "No one speaks English and everything's broken." -- T. Waits
> "Tom Waits would have made a decent sysadmin." -- M. Jinks
>
> _______________________________________________
> Redhat-list mailing list
> [EMAIL PROTECTED]
> https://listman.redhat.com/mailman/listinfo/redhat-list
>
> _______________________________________________
> Redhat-list mailing list
> [EMAIL PROTECTED]
> https://listman.redhat.com/mailman/listinfo/redhat-list
_______________________________________________
Redhat-list mailing list
[EMAIL PROTECTED]
https://listman.redhat.com/mailman/listinfo/redhat-list
