ftp.nice.ch/pub/next/unix/disk/ufs-NeXT.0.3.s.tar.gz#/linux/fs/ufs

Makefile
 
README
 
dir.c
[View dir.c] 
file.c
[View file.c] 
inode.c
[View inode.c] 
namei.c
[View namei.c] 
ufs.h
[View ufs.h] 

README

This is an ALPHA patch for adding BSD ufs filesystem support (readonly for
now) to linux. The patch was made against kernel 1.1.55.

The patch includes additional 1 line to config.in, but as my config.in is
probably different than yours, this might go wrong. Not a big deal.

My setup of FreeBSD lives in its own partition, which is sub-divided by the
BSD disklabel; therefore the partition recognition code (genhd.c) now
handles disklabel-in-partition, similar to the way extended partitions are
handled. This, however, have two problems. First, the number of total
LOGICAL (i.e. inside-BSD or inside-extended) partitions (hdx5 and up) is
small, and if you have both extended and BSD, or several BSDs, some
partitions may not be accessible.

The other problem is the assignment of device numbers; If your BSD partition
is processed before the DOS-like extended partition, the logical partitions
read from the extended partition will be shifted to higher numbers (or
ignored, if max. number of logical partitions exceeds).

This will have to be solved sometime, but implementing read/write is more
important for me. It is also possible that the disklabel-treatment will be
taken off the bootup code, and placed as a ioctl which will have to be
called before mounting.

There are some other cases that are not currently considered, such as an
installation with a standalone disklabel, i.e. with no dos-like partition
table. Also, cases where the fragment size isn't 1024 or 512 aren't
supported (though it MIGHT work).

This patch comes with ABSOLUTLY NO WARRANTY.



I would like to hear some comments about this, as I don't really know much
about possible bsd installations.

Yossi Gottlieb  (yogo@math.tau.ac.il)

These are the contents of the former NiCE NeXT User Group NeXTSTEP/OpenStep software archive, currently hosted by Netfuture.ch.