Path: utzoo!utgpu!jarvis.csri.toronto.edu!clyde.concordia.ca!uunet!snorkelwacker!usc!brutus.cs.uiuc.edu!samsung!aplcen!haven!udel!mmdf From: postmaster@att-in.att.com Newsgroups: comp.os.minix Subject: (none) Message-ID: <8391@nigel.udel.EDU> Date: 16 Jan 90 18:11:00 GMT Sender: mmdf@udel.EDU Lines: 58 Mail to `att.att.com!attmail!mhs!envoy!ics.test/pn=_test_group' alias `att!attmail!mhs!envoy!ics.test/pn=_test_group' from 'VM1.NoDak.EDU!INFO-MINIX%UDEL.EDU' failed. The error message was: destination unknown or forwarding disallowed The message began: Received: from NDSUVM1.BITNET by VM1.NoDak.EDU (IBM VM SMTP R1.2.1MX) with BSMTP id 2971; Tue, 16 Jan 90 11:07:34 CST Received: from NDSUVM1.BITNET by NDSUVM1.BITNET (Mailer R2.03B) with BSMTP id 2954; Tue, 16 Jan 90 11:07:32 CST Date: Tue, 16 Jan 90 13:29:28 GMT Reply-To: INFO-MINIX%UDEL.EDU@VM1.NoDak.EDU Sender: Minix operating system Comments: Warning -- original Sender: tag was info-minix-request@UDEL.EDU From: Aubrey McIntosh Subject: Re: disk geometry. Geometry Sector. Comments: To: info-minix@udel.edu To: Multiple recipients of list MINIX-L In article lloyd@comp.vuw.ac.nz (Lloyd Parkes) writes: >In article <7866@nigel.udel.EDU> Peter_Van_Epp@cc.sfu.ca writes: > >The problem of finding out the geometry of a disk with no file system >on it is very hard (if you don't want to thrash the drive). One >solution, might be start the sector numbering (on track 0 anyway) from >some number other than zero. Different start numbers would indicate >different disk geometries, and you would avoid those deadly seek to >track 79 :-) This of course raises other problems. It is trivial (a mathematical term meaning either totally obvious to everybody, or incomprehensible to the author as well) to change the interleave. It it more interesting to detect the change. However, a very similar idea might work. The sector number and size are given to the NEC controller on a sector by sector basis during format time. It is possible to make a sector number, say 0, that is a small sector and not a usual one. Such a sector could be explicitly read by the boot software, and would otherwise be unobtrousive. The geometry sector would not be copied or written by dd, tar, mkfs, ... under the existing code. In those cases where another sector will not fit due to some circumstance of controller/drive/media brand, one of the standard sectors could be replaced with the sentinal sector, and the bit map told that it is bad. This resultant disk could be used interchangeably with other operating systems, since the NEC reads until either a sector id match occurs, or a double index hole detect occurs. Funny and/or out of sequence numbers are (typically--exclusive of sophisticate track buffer code) no problem. One of the features on the list-to-implement in my format program is a "read existing data and keep it" feature. It would be straightforward to extend that to "add a disk geometry sector" as well. -- Aubrey McIntosh comp.os.minix, comp.lang.modula2, soc.culture.celtic Austin, TX 78723 1-(512)-452-1540 (v)