Just a thought that popped into my head

Jesse Tie-Ten-Quee highos at linuxfromscratch.org
Tue May 7 04:33:19 PDT 2002


Yo,

On Tue, May 07, 2002 at 04:19:34AM +0100, Ian Molton wrote:
> have you run benchmarks? perhaps you might revert to 2.95.3...

Who runs benchmarks on a normal little box?  It isn't such a big deal,
now is it?  If it works, it works.  From what i've read, the GCC Team
considers a %15 slower (for certain things) "acceptable" for a gcc 3.x
release, so really...

> besides, is there a good reason to ditch 2.95.3?

The C++ support for one, afiak.  I know many projects welcomed that.

However, *you* do not have to ditch it.  This is not what the issue is
about.  This is about the lfs-book staying current with what is
_currently stable_, which gcc3 is.

> reality check time... who has a genuine problem with 2.95.x?

Who has a genuine problem with 3?  And i'm not talking about what the
developers are posting on there mailing lists.  We are not them and they
are not us.  My personal view of the GCC Team is a joke, mostly because
they never hit there own release dates.. at this point they should just
pick up the E Team's view of just releasing it "When it's ready."
instead of always trying to hit a date they never will.

I've been hearing nothing but very good stories about gcc3 for the last
couple of months... and if you read what Gerard has posted, he WILL
consider not moving up, however he needs PROOF.  Not just what someone
is saying or reporting, he needs to be able to look at it and say "Yup,
this is bad, we better not upgrade".

> the compiler is the absolute foundation of the system, and GCC 3.x is KNOWN to be flakey, even to the people writing it, at the moment!

That has never stopped us before ;)

> we should wait until the GCC people tell us GCC 3.x is ready for prime time.

And that would problably be never, my friend.  Not from my experience
with the GCC Team ;)

Anyways, i've tried to stall the gcc3 upgrade over the last couple of
months, and I can no longer stand in the way.  I settled for waiting
untill the 3.1 release was available, then upgrading, it took alot of
offer just to get that settlement out of some ppl ;)

If you want to keep trying to change people's minds, then i suggest you
start getting some real proof, and not just saying that the developers
are not "happy" with the present state the code is in.  Hey, 3.1 isn't
released yet, so id hope they aren't happy ;)

Anyways...

-- 
Jesse Tie-Ten-Quee  ( highos at linuxfromscratch dot org )
-- 
Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe lfs-dev' in the subject header of the message



More information about the lfs-dev mailing list