Subject: 1.6_STABLE -> 1.6K broken?
To: None <port-macppc@netbsd.org>
From: =?iso-2022-jp?B?GyRCSy0xShsoQiAbJEJDIz9NGyhC?= <toyonaga@rr.iij4u.or.jp>
List: port-macppc
Date: 11/11/2002 02:19:15
I'm having the difficulties in shifting from 1.6_STABLE to CURRENT on my
recentry purchased iBook (Dual USB,G3/700Mhz,640M ram).

I installed 1.6-RELEASE by ftp and then grabbed netbsd-1-6 branch source
from local CVS repositorye, which I cvsup in another machine.

Building 1.6_STABLE on 1.6-RELEASE is as easy as I used to do on i1386 and
it properly boots.

Then, I did:
  mkdir /usr/current
  (cd /usr/src; ./build.sh -O /usr/obj -T /usr/tools -D /usr/current)
  (cd /usr/src; ./build.sh -O /usr/obj -T /usr/tools -D /usr/current -r -u)
  (cd /usr/current/usr; cvs -d /home/ncvs checkout -P src)
  chroot /usr/current
  cd /usr/src                               ( actually /usr/current/usr/src)
  ./build.sh -T /usr/tools -O /usr/obj -D /usr/dest -t
  ./build.sh -T /usr/tools -O /usr/obj -k CURRENT
                            (CURRENT = GENERIC - ALTIVEC + OFB_ENABLE_CACHE)
  exit                                          (from chroot'ed environment)
  mv /usr/current/usr/src/sys/arch/macppc/CURRENT/netbsd /netbsd.CURRENT
  shutdown -r now 
  boot hd:9,\ofwboot.xcf hd:10,/netbsd.CURRENT
  chroot /usr/current
  (cd /usr/src; ./build.sh -O /usr/obj -T /usr/tools -D /usr/dest)
  (cd /usr/src; ./build.sh -O /usr/obj -T /usr/tools -D /usr/dest -r -u)
  exit

Now I have a fresh current dist in /usr/current/usr/dest. 
But chroot /usr/current/usr/dest only leaves csh.core in /usr/current/usr/dest
saying "illegal instruction".

Chrooting to /usr/current/usr/dest with /rescure/csh as second parameter is
okay, but dynamically linked executables never work, just dumps core...

Why I did such staged build is because I brew my system just doing
./build.sh -O /usr/obj -T /usr/tools on 1.6_STABLE with current- src.
In that case, I couldn't go into multiuser with a lot of same
 "illegal instruction" s. 

Daily builds from releng.netbsd.org also shows simillar results.

Any advice?