Hi Riccardo,

> In my opinion you are casting dark shadows... wen is "2.0" about due ?

When core decides which ports to drop in order to release it, as I
understand it. We're up to RC5, and we can't expect perpetual RCs.

> from my PoV the only problem that got in addition to 1.6.2 was the
> sync stuff that we could see as "almost fixed".
> On my machine, 1.6.2 had not a working X, had no usable mc0 and boot
> problems. That is "no new problems" so 2,0 wouldn't be a worse
> experience than 1.6.2 was.

Well, the MP kernel panics regularly (I can't build userland on it with
less than 256M), the mount_mfs issue causes kernel panics, there appears
to be an issue in disklabel that will pick up previous NetBSD installs
even if the disk was reformatted HFS in between, grackle won't properly
attach PCI cards (patch available, needs committing), if there is more
than one disk with a NetBSD install the code picks up the lowest SCSI ID
for booting even if the higher one is the booted device, some of the
devices behind bridges are not getting initialized properly, and the
bootloader still has failures on some boxes.

I found all of the above independently of any review of PRs in less than
three weeks of using RC4. I don't think I'm being negative when I say
let's stop and assess where we are and proceed as we can. I'd much
rather come out later than other ports than release with the above bugs
in it and have people evaluate it.

tim