Subject: Re: various changes on INSTALL and iso generation
To: Chavdar Ivanov <email@example.com>
From: Manuel Bouyer <firstname.lastname@example.org>
Date: 02/27/2007 14:22:07
On Tue, Feb 27, 2007 at 01:06:46PM +0000, Chavdar Ivanov wrote:
> On 27/02/07, Manuel Bouyer <email@example.com> wrote:
> >On Mon, Feb 26, 2007 at 11:27:36PM -0800, Erik Berls wrote:
> >> [cc: - port-i386]
> >> Excelllent. I was considering making ACPI and MP default for
> >> port-amd64. With the addition of -NOMP and -NOACPI kernels. The
> >> mailing list seemed to errupt in a thread about ACPI breaking on a
> >> large number of machines, so I aborted, and then didn't have time to
> >> revisit.
> >> Can we get MP added as a default?
> >It's already available from the list of kernels to install, I'm not
> >sure if it's a good idea to make it the default. The performance hit
> >is noticeable on UP machines. Ideally sysinst should be able to make
> >SMP or UP the default after detecting the number of CPU, but that's
> >another story.
> >> How much of these fixes can get pulled up to netbsd-4? (eg. if we
> >> are able to patch for the dell, etc..)
> >I intend to get these in netbsd-4. But I have some more changes to do to
> >iso build infrastructure first.
> I finished build.sh release for 4.99.13 a few hours ago on amd64 with
> the diff applied. The resulting ISO booted fine and recognized
> everything to be recognized, but could not be used for installation as
> there was no MAKEDEV.subs present in /dev, just MAKEDEV (that was the
> message I got when I tried ./MAKEDEV all in /dev). Have I done
Not sure. I did a complete i386 install, but I just booted amd64.
Unfortunably qemu x86_64 fails to boot this image, so I need to test
an install on a real machine.
Manuel Bouyer <firstname.lastname@example.org>
NetBSD: 26 ans d'experience feront toujours la difference