From mboxrd@z Thu Jan  1 00:00:00 1970
From: tlaronde@polynum.com
Date: Wed, 23 May 2007 20:35:56 +0200
To: Fans of the OS Plan 9 from Bell Labs <9fans@cse.psu.edu>
Message-ID: <20070523183556.GA2037@polynum.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.4.2.2i
Subject: [9fans] MBR messed up on installation
Topicbox-Message-UUID: 712a8de2-ead2-11e9-9d60-3106f5b1d025


Hello,

Just a note for the record (I think I will have the time tomorrow to
track down the problem and give explanations/patches) for the ones
installing Plan 9 on a i386 with a NetBSD partionned disk. 


When finishing the install, and setting (there is not the choice to "do
nothing") the active partition to Plan 9, apparently all the slots are
rewritten in the table, and the chs and xstart for previous alien
partitions are recomputed (modified), leading obviously to the
impossibility to launch NetBSD and to find its disklabel.

	At first look it seems that Plan 9 recomputes the beginning of the
	partition to match a cylinder border using BIOS values (number of
	sectors by cylinder) while NetBSD uses a disklabel, kernel provided
	value for the same task:
	
$ fdisk wd0

Disk: /dev/rwd0d
NetBSD disklabel disk geometry:
cylinders: 38792, heads: 16, sectors/track: 63 (1008 sectors/cylinder)
total sectors: 39102336

BIOS disk geometry:
cylinders: 1024, heads: 255, sectors/track: 63 (16065 sectors/cylinder)
total sectors: 39102336

Partition table:
0: Plan 9, or Theos spanned (sysid 57)
    bootmenu: Plan9
    start 63, size 2232972 (1090 MB, Cyls 0-139), Active
1: NetBSD (sysid 169)
    start 2241792, size 14329728 (6997 MB, Cyls 139/139/1-1031/135/1)
2: NetBSD (sysid 169)
    bootmenu: NetBSD
    start 16571520, size 8192016 (4000 MB, Cyls 1031/135/1-1541/117/1)
3: <UNUSED>
Bootselector enabled, timeout 10 seconds.

i.e. (for partition 1) 2241792 % 1008 == 0 

while Plan 9 gives 

xstart = 139 * 16065 = 2233035  for the same partition (p2)


I will verify this later (a priori, there is a non identical conversion

xstart -> chs
chs -> xstart

and probably the only solution is to take as is the values for the
partitions about which Plan 9 fdisk did not modify the _geometry_.
)

Note 2: could someone update the Wiki page about the installation to
note that the minimum of 300 Mbytes is, for the current distribution and
replica settings, for the sole fossil partition, and not for the whole 
chunk dedicated to Plan 9?

Cheers,
-- 
Thierry Laronde (Alceste) <tlaronde +AT+ polynum +dot+ com>
                 http://www.kergis.com/
Key fingerprint = 0FF7 E906 FBAF FE95 FD89  250D 52B1 AE95 6006 F40C