[comp.unix.xenix.sco] Hard disks with > 1024 Cylindars

rompub@gpu.utcs.utoronto.ca (Mark Dornfeld) (08/31/90)

I have purchased a Wren VI Harddisk for installation under SCO Xenix
and/or SCO Unix ODT.  The disk has 1747 cylindars, 7 head and 54
sectors/track.  I am using an Everex ESDI controller that understands
what this disk is all about and has routines to handle it.

The SCO ODT documentation suggests that such a disk is compatible as
long as the root partition doesn't cross over the 1024 cylindar
boundary.  Now, my limited understanding of disk geometry tells me
that is not an easy task.  

Has anyone had any experience with this or similar drives under Xenix
or Unix.  Can it be used reliably?

Mark T. Dornfeld
Royal Ontario Museum
Toronto, Ontario

utgpu!rom!mark  or utgpu!rompub

Please reply via email and I will summarize and post.

Thanks for any help in advance.

chip@chinacat.Unicom.COM (Chip Rosenthal) (09/01/90)

In article <1990Aug31.022148.5775@gpu.utcs.utoronto.ca>
	utgpu!rom!mark, utgpu!rompub writes:
>The disk has 1747 cylindars, 7 head and 54 sectors/track.  [...]  The SCO ODT
>documentation suggests that such a disk is compatible as long as the root
>partition doesn't cross over the 1024 cylindar boundary.

The problem is that the peecee BIOS can't grok more than 1024 cylinders.
Until UNIX is up and flying, you are a slave to the limitations of the
BIOS.  One of the activities during this stage is to load in /boot and
then /unix.  Therefore, they must be accessible by BIOS.  If your root
partition is bigger than 1024 cylinders then /boot or /unix could be
sitting in a region which is inaccessible.  So, limiting /dev/root to
1024 cylinders guarantees they can be loaded.

Some ESDI controllers, such as the WD1007, do remapping, and present a
different disk geometry to the system.  Thus, as far as the system is
concerned the disk *is* less than 1024 cylinders.

As a practical matter, if you use a reasonable disk divvy, you generally
aren't going to have filesystems this large.

Oh yeah, why are we talking about unix in comp.unix.xenix.sco anyway?
-- 
Chip Rosenthal  <chip@chinacat.Unicom.COM>
Unicom Systems Development, 512-482-8260 
Our motto is:  We never say, "But it works with DOS."

romkey@asylum.SF.CA.US (John Romkey) (09/02/90)

In article <1990Aug31.022148.5775@gpu.utcs.utoronto.ca> utgpu!rom!mark, utgpu!rompub writes:
>Has anyone had any experience with this or similar drives under Xenix
>or Unix.  Can it be used reliably?

I'm using a 1658 cylinder, 15 head ESDI disk with a WD1007SV-?? (can't
remember those letters right now) disk controller under SCO Xenix with
no problems; I can access the entire disk (no geometry remapping or
anything). I hadn't heard about the root partition restriction, but
that should be relatively easy to get around. I suspect the problem is
that /boot reads in /xenix (or /unix) using BIOS routines, which
probably won't be able to go beyond 1024 cylinders, so you have to
guarantee that /xenix or /unix is entirely within the first 1024
cylinders. So, just make the root partition be less than 1024
cylinders long and you're set. Put other stuff like user accounts in
partitions like /u.
-- 
			- john romkey
USENET/UUCP: romkey@asylum.sf.ca.us	Internet: romkey@ftp.com
King Kong died for your sins.

romkey@asylum.SF.CA.US (John Romkey) (09/02/90)

In article <1535@chinacat.Unicom.COM> chip@chinacat.Unicom.COM (Chip Rosenthal) writes:
>Some ESDI controllers, such as the WD1007, do remapping, and present a
>different disk geometry to the system.  Thus, as far as the system is
>concerned the disk *is* less than 1024 cylinders.

I got bitten by a remapping problem with the WD1007 once. It can remap
up to a maximum of 1024 cylinders, 64 sectors per track and 16 heads.
That multiplies out to a maximum of 512MB, which is sad when you're
using a 660MB disk like I was. Then I discovered that Xenix could deal
with the controller without mapping and still access the entire disk,
and became very, very happy.
-- 
			- john romkey
USENET/UUCP: romkey@asylum.sf.ca.us	Internet: romkey@ftp.com
King Kong died for your sins.

art@pilikia.pegasus.com (Art Neilson) (09/03/90)

In article <1990Aug31.022148.5775@gpu.utcs.utoronto.ca> utgpu!rom!mark, utgpu!rompub writes:
>I have purchased a Wren VI Harddisk for installation under SCO Xenix
>and/or SCO Unix ODT.  The disk has 1747 cylindars, 7 head and 54
>sectors/track.  I am using an Everex ESDI controller that understands
>what this disk is all about and has routines to handle it.
>
>The SCO ODT documentation suggests that such a disk is compatible as
>long as the root partition doesn't cross over the 1024 cylindar
>boundary.  Now, my limited understanding of disk geometry tells me
>that is not an easy task.  
>
>Has anyone had any experience with this or similar drives under Xenix
>or Unix.  Can it be used reliably?

You can either split your system into separate filesystems such that
root doesn't cross the boundary, or turn on sector mapping in your 
controller.  You can do this from DOS by going into debug and giving
the command g=c800:5, note that some controllers may have their ROM
at a different location, consult your controller docs.  You should
get some kinda menu with a choice for setting special options, one
of the options should enable the controller to map the sectors such
that it presents < 1024 cylinders to your OS.
-- 
Arthur W. Neilson III		| ARPA: art@pilikia.pegasus.com
Bank of Hawaii Tech Support	| UUCP: uunet!ucsd!nosc!pegasus!pilikia!art

bote@csense.uucp (John Boteler) (09/07/90)

From article <1535@chinacat.Unicom.COM>, by chip@chinacat.Unicom.COM (Chip Rosenthal):
> 
> Oh yeah, why are we talking about unix in comp.unix.xenix.sco anyway?

I was just gonna ask why we have comp.unix.xenix.sco when that's the
only Xenix anybody remembers nowadays.

I have been running an 1177 cylinder disk under Xenix for years now,
no problem, first with a DTC7287 controller, and now with a
WD1006VSR2.

I think the WD is not making my floppy interface happy: it locks
the system under Xenix when simultaneous floppy and hard disk
activity occurs while the system is set to full speed. :(

Udderwise, it works fine with > 1024 cylinders, despite SCO's
protestations to the contrary.

-- 
John Boteler   bote@csense.uucp           {uunet | ka3ovk}!media!csense!bote
SkinnyDipper's Hotline: 703-241-BARE | VOICE only, Touch-Tone(TM) signalling

root@ozdaltx.UUCP (root) (09/10/90)

In article <1990Sep7.052138.495@csense.uucp>, bote@csense.uucp (John Boteler) writes:
> I have been running an 1177 cylinder disk under Xenix for years now,
> no problem, first with a DTC7287 controller, and now with a
> WD1006VSR2.

After a brief conversation with a SCO service rep the other day, He
explained that the 1024 limit is only on the boot/active partition.
If your primary is bigger than 1024, then another partition/file
system should be created.  Secondarys appearently don't suffer this
limit.  I just added a secondary with 1314 cylinders and it singing
along quite merrily.  This is on a MFM interface.
         scotty
         root@ozdaltx

davidsen@sixhub.UUCP (Wm E. Davidsen Jr) (09/13/90)

In article <1990Sep7.052138.495@csense.uucp> bote@csense.uucp (John Boteler) writes:

| I think the WD is not making my floppy interface happy: it locks
| the system under Xenix when simultaneous floppy and hard disk
| activity occurs while the system is set to full speed. :(

  I bet you haven't installed the SLS for that. Although it's called the
WD1007 fix, it really works with the 1006, too. I don't have the list
handy, from memory xnx133, but it's the one which says WD1007.
-- 
bill davidsen - davidsen@sixhub.uucp (uunet!crdgw1!sixhub!davidsen)
    sysop *IX BBS and Public Access UNIX
    moderator of comp.binaries.ibm.pc and 80386 mailing list
"Stupidity, like virtue, is its own reward" -me