On Thu, Jan 24, 2013 at 03:27:12PM +0100, Ingo J?rgensmann wrote:
> Am 24.01.2013 um 10:09 schrieb "Christian T. Steigies" <c...@debian.org>:
> 
> >> You should preferrably use either LVM or BTRFS snapshotsto speed up 
> >> deinstallation of build-dep packages. I've chosen LVM because I'm 
> >> unfamiliar with BTRFS, which is the most recommended option for handling 
> >> chroots. 
> > I am a bit familar with LVM. Can I use this on one partition only instead of
> > the whole disk?
> 
> Sure. Just choose a partition for LVM and do a pvcreate /dev/sdXY, pvcreate & 
> lvcreate. 

And the idea is with LVM I can quickly create snapshots for building, user
chroots, etc? How much space should a chroot then receive, how much should I
leave for the snapshots? The partiton I would use has 160GB.
 
> > So far I always used chroot in standard directories, but
> > debootstrap failed in the end with this:
> > W: Failure trying to run: chroot /data/chroot/buildd/sid mount -t proc proc 
> > /proc
> > W: See /data/chroot/buildd/sid/debootstrap/debootstrap.log for details
> > chroot: failed to run command 'mount': No such file or directory
> > I dimly remember that I have to mount the proc directory on the host system,
> > but this error message is not very helpful.
> 
> Yes, I had the same error, but I don't remember how I worked around it. :-/

deboostrap is slooooow... it took four hours to get to the same point, and
fail again. I had mounted proc and sysfs before running debootstrap, but I
think the actual problem is that some libraries are missing in the chroot
system:

root@kullervo:~>ldd /data/chroot/buildd/sid/bin/sh  
        libc.so.6 => /lib/m68k-linux-gnu/libc.so.6 (0xc0005000)
        /lib/ld.so.1 (0xd0000000)

neither of them is available in the chroot. Do I need to add those packages
manually? My debootstrap line is this:

debootstrap \
 --keyring=/usr/share/keyrings/debian-ports-archive-keyring.gpg \
 --variant=buildd sid \
 /data/chroot/buildd/sid http://ftp.debian-ports.org/debian/ 
 
> > [    0.000000] Ignoring memory chunk at 0x7000000:0x1000000 before the 
> > first chunk
> > [    0.000000] Fix your bootloader or use a memfile to make use of this 
> > area!

> Hmm, I guess kullervo has some additional onboard Fast-RAM (ZIP Chips). That 
> can be up to 16 MB RAM. When you substract this amount from 128M - kernel 
> space it would make up to 20 MB, resulting in roughly 107 MB usuable RAM. But 
> why should amiboot or the kernel substract the onboard Fastmem from the 
> accelerator card memory? 
> 
> There seems to be some information about amiboot on 
> ftp://ftp.uni-erlangen.de/pub/Linux/680x0/v2.0/boot/amiboot-4.0.README 
> including information about memfile. I guess Geert can better comment on 
> amiboot questions.

I will have a look later. But the kernel is ignoring some memory, maybe it
is this. Or the kernel uses too much, but this is only 5MB, the video uses 640K:

[    0.000000] Memory: 109452k/109452k available (2740k kernel code, 2372k 
data, 124k init)
[    0.710000] fb0: Amiga ECS frame buffer device, using 640K of video memory


Christian


-- 
To UNSUBSCRIBE, email to debian-68k-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130124151803.gb17...@chumley.earth.sol

Reply via email to