Re: Performance 4.x vs. 6.x (was: e: [fbsd] HEADS UP: FreeBSD 5.3, 5.4, 6.0 EoLs coming soon)

--- Dan Lukes <dan@xxxxxxxxx> wrote:

Danial Thom wrote:
The right thing to do is to port the SATA
and new NIC support back to 4.x and support
4.x is far superior on a Uniprocessor system
FreeBSD-5+ may be an entire re-write away
ever being any good at MP. Come to terms with
PLEASE, because it is the case and saying
otherwise won't change it.

Despite I'm initiator of this way of
discussion (in security list), I
can't agree with you. No way.

You are not allowed to tell to someone working
as volunteer several
months on something that the best way is
rollback all work and start
from scratch. Despite of your complaints are
competent or not. You
totally miss the right time for this type of
complain. It's too late now.

6.x is not crap in any way. It has some
problem, even after many months
of development, but it can be resolved if
volunteers decide to use it's
power to polish previously implemented code.
Current 6.x is better in
many parameters than 4.x. Well, some important
parameters are worse, but
correct decision is improve them, not rollback
all work.

I voted against premature EOLing of 4.x, but
returning to FreeBSD 4.x
is not acceptable way in any way - at least
because it's the DragonBSD's
nest now.

I didn't say to roll back all of the work. I said
to support 4.x as a UP solution, and 6,7 or
whatever as what it is now. 5+ will never be as
good as 4.x UP, and many networking applications
such as firewalls and routers simply will never
be able to scale to utilize MP anyway. You had
the best damn UP OS in the world, why not
continue to support it as so while you try to
figure out MP?


Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
freebsd-stable@xxxxxxxxxxx mailing list
To unsubscribe, send any mail to "freebsd-stable-unsubscribe@xxxxxxxxxxx"