From owner-freebsd-questions Wed Sep 2 15:58:23 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id PAA21690 for freebsd-questions-outgoing; Wed, 2 Sep 1998 15:58:23 -0700 (PDT) (envelope-from owner-freebsd-questions@FreeBSD.ORG) Received: from dan.emsphone.com (dan.emsphone.com [199.67.51.101]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id PAA21680 for ; Wed, 2 Sep 1998 15:58:18 -0700 (PDT) (envelope-from dan@dan.emsphone.com) Received: (from dan@localhost) by dan.emsphone.com (8.9.1/8.9.1) id RAA11793; Wed, 2 Sep 1998 17:56:52 -0500 (CDT) Message-ID: <19980902175652.A11500@emsphone.com> Date: Wed, 2 Sep 1998 17:56:52 -0500 From: Dan Nelson To: Brett Glass , sab@seanet.com Cc: Doug White , questions@FreeBSD.ORG Subject: Re: MSDOS extended partitions and "slices" References: <199808080608.AAA16222@lariat.lariat.org> <199808021131.FAA12204@lariat.lariat.org> <199808080608.AAA16222@lariat.lariat.org> <199809021901.MAA21131@two.sabami.seaslug.org> <199809022113.PAA00535@lariat.lariat.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.94.2i In-Reply-To: <199809022113.PAA00535@lariat.lariat.org>; from "Brett Glass" on Wed Sep 2 15:09:05 GMT 1998 X-OS: FreeBSD 2.2.7-STABLE Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG In the last episode (Sep 02), Brett Glass said: > At 12:01 PM 9/2/98 -0700, Scott Blachowicz wrote: > >Brett Glass wrote: > >> But.... Waitaminnit. If you have an extended DOS partition with some > >> number of logical DOS drives within it, you should REALLY see: > >> > >> C: wd1s1 > >> Extended DOS partition: wd1s2 > >> D: wd1s2a > >> E: wd1s2b > >> F: wd1s2c > >> Third partition (FreeBSD, Linux, whatever): wd1s3 > >> Fourth partition: wd1s4 > > > >Wait a minute...I thought the logical DOS drives within the extended > >partition corresponded to slices 5 & up....as in... > > > > C: wd1s1 > > Extended DOS partition: wd1s2 > > D: wd1s5 > > E: wd1s6 > > F: wd1s7 > > Third partition (FreeBSD, Linux, whatever): wd1s3 > > Fourth partition: wd1s4 > > You see the latter. My point is that this is not logical. Sure it's logical. FDISK slices are always numbers; disklabel partitions are always letters. wd1s1 .. wd1s4 are the four primary fdisk slices. Any or all of them MAY be a container (i.e. extended partition type). If any of them are, then they are scanned for valid partition entries (including more nested extended partitions), and any of those will appear at wd1s5, s6, etc. The kernel sets an arbitrary limit of 32 of these slices. So say you had a mythical 14G IDE drive partitioned with FAT16 DOS drives: +--------+-----------------------------------------------------+ | C: | extended partition | +--------+- - - - + - - - -+- - - - + - - - - - - - - - - - - -+ | D: | E: | F: | extended part. | +--------------------------+ - - - - - - - - - - - - -+ | G: | H: | I: | +--------------------------+ (boy, I respect Terry more now. Drawing ascii-art isn't easy :) In your naming scheme, what would drive 'G:' be? wd1s2da ? FreeBSD would name them like this (after looking at the code in diskslice_machdep.c; I could be wrong): wd1s1 C: wd1s2 extended partition container #1 wd1s3 empty wd1s4 empty wd1s5 D: wd1s6 E: wd1s7 F: wd1s8 extended partition container #2 wd1s9 G: wd1s10 H: wd1s11 I: -Dan Nelson dnelson@emsphone.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message