Date: Thu, 28 Dec 1995 21:26:26 +0100 From: Poul-Henning Kamp <phk@critter.tfs.com> To: Michael Smith <msmith@atrad.adelaide.edu.au> Cc: grog@lemis.de, cosmos@misery.bssc.org, hackers@FreeBSD.org, jack@cdrom.com Subject: Re: HELP!!! THIS IS AN EMERGENCY (fwd) Message-ID: <16653.820182386@critter.tfs.com> In-Reply-To: Your message of "Fri, 29 Dec 1995 02:15:38 %2B1030." <199512281545.CAA03130@genesis.atrad.adelaide.edu.au>
next in thread | previous in thread | raw e-mail | index | archive | help
> The "best" solution to the basic problem (no Jordan, no wolves this time 8)
> would be better detection/interoperation with BIOS extenders. This is
> kinda problematic though, as none of the deep-down disk hackers consider IDE
> worthy of their spit, and the technical aspects of the problem are
> pretty hairy.
As the one who spent a lot more than I liked, of time I didn't have, on
getting this particular area of sysinstall working, I kind of resent that
remark, quite a lot in fact :-(
I am on the other hand painfully aware of how badly we fare in many
cases, so I will let it pass for now :-)
I am also on the ata/atapi mailing-list and I can comfortably say that
this entire area is so impossible to handle intelligently and correctly
in all cases, that I'm seriously tempted to not even touch it again, ever.
Now just the other day Jordan prodded me with a big stick to look at
it again, and I guess like the sucker I am, I will...
For you amusement and education, here is the list of stuff you need
to test on to make sure you didn't goof up.
1) a ST506 disk, (remember the 20Mb kind ?)
2) an ESDI disk with > 1024 cylinders
3) an IDE disk with < 1024 cylinders
4) an IDE disk with > 1024 cylinders
* 5) a SCSI disk < 1GB
* 6) a SCSI disk > 1GB, < 2GB
7) a SCSI disk > 2GB, < 4GB
* 8) a SCSI disk > 4GB, < 8GB
* 9) a NCR SCSI controller
10) an AHA1542 controller
11) an AHA1742 controller
12) an EISA system for 11)
* 13) a BT545 controller
* 14) a BTxxx controller
* .... You get the idea...
* A) a copy of MSDOS < ver 3.0
* B) a copy of MSDOS ver 3.x
* C) a copy of MSDOS ver 5.x
D) a copy of MSDOS ver 6.x
* E) one or two different OS/2 version
* F) a copy of NT
* G) a copy of Solaris
H) a copy (seven actually :-) of Linux
I) a copy of Ontrack Diskmanager version 6.X
* J) a copy of Ontrack Diskmanager version 7.X
* .... You get the idea...
Well, the * marked Items was what I didn't have available last time,
and several more are unavailable for me now.
Not to mention the time to shuffle it around all the time.
As you can gather from the above, we will never be able to make a
"Do What I Mean" installation in all cases.
No way!
Never!
It doesn't improve too much that people send error reports which
essentially say "Hey it fucked my harddisk over, die you sucker!!!"
Correlating that description to the code is beyond the very limited
supernatural forces I have been given.
Sometimes I wonder why I do this for fun...
If you have a configuration where we fail, please stay calm, steal
you younger brothers computer or something and contact us so we can
solve this problem.
I personally have 12+ years of OS installation, and have learned that
if I have anything on the disks I care about, I will back it up first,
and I will read any information from the install program very very
carefully, with a very paranoid attitude.
And I generally still tend to keep the BRS (Big Red Switch) inside panic
range all the time.
--
Poul-Henning Kamp | phk@FreeBSD.ORG FreeBSD Core-team.
http://www.freebsd.org/~phk | phk@login.dknet.dk Private mailbox.
whois: [PHK] | phk@ref.tfs.com TRW Financial Systems, Inc.
Future will arrive by its own means, progress not so.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?16653.820182386>