login
Header Space

 
 

Re: scsi disk partitioning and bootstrap questions

Score:
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
Date: Saturday, August 29, 1992 - 8:29 am

In article <1992Aug28.203237.9239@osf.org> ede@kadabra.osf.org (Ted Ede) writes:

The limit is caused by the number of bits in the partition table which
are available for storing sectors per track: 6 bits.  Therefore a
sector number >63 is not addressable.  The whole problem of getting a
new partition table format understandable by all operating systems
is going to be a major headache in the future.


Your partition table is rather strange, but never mind!  When I wrote
fdisk I used 'begin' for the start-of-partition encoded in terms of
geometry, and 'start' for the actual start-of-data encoded in terms
of sector number.  These two numbers need not be the same, though for
Linux fs and swap partitions any difference is simply wasted.


You can use any partition for lilo.  Shoelace is becoming more problematic
all the time, and is better avoided.


If you use lilo, you can start root with any sector on the disk, regardless
of cylinder boundaries.


Shoelace was a holdover from MINIX: it was adapted to Linux to give us the
possiblilty of booting from the hard disk, but it could not be compiled
under Linux, was very tempermental, and is probably not going to work
indefinitely.  Shoelace requires that you use the Minix file system, while
lilo will work with anything.  Lilo also uses a great deal less disk space.


     -- Owen
     LeBlanc@mcc.ac.uk
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
Re: scsi disk partitioning and bootstrap questions, A. V. Le Blanc, (Sat Aug 29, 8:29 am)
speck-geostationary