27 messages in org.freebsd.freebsd-stableRe: *HEADS UP* Correction to previous...| From | Sent On | Attachments |
|---|---|---|
| Mike Smith | Mar 9, 1998 12:52 am | |
| Bruce Evans | Mar 9, 1998 2:19 am | |
| Gary Schrock | Mar 9, 1998 11:16 am | |
| Bruce M. Walter | Mar 9, 1998 1:58 pm | |
| Mike Smith | Mar 9, 1998 2:10 pm | |
| Bruce Evans | Mar 9, 1998 6:01 pm | |
| Mike Smith | Mar 9, 1998 6:55 pm | |
| Bruce Evans | Mar 9, 1998 7:42 pm | |
| Joao Carlos Mendes Luis | Mar 9, 1998 8:37 pm | |
| Bruce Evans | Mar 9, 1998 9:17 pm | |
| Luoqi Chen | Mar 10, 1998 12:27 am | |
| Bruce Evans | Mar 10, 1998 2:00 am | |
| Mike Smith | Mar 11, 1998 3:24 am | |
| Gary Schrock | Mar 11, 1998 9:47 am | |
| David E. O'Brien | Mar 13, 1998 6:50 pm | |
| Bruce Evans | Mar 13, 1998 8:36 pm | |
| David E. Tweten | Mar 13, 1998 9:16 pm | |
| Mike Smith | Mar 13, 1998 10:40 pm | |
| David E. O'Brien | Mar 13, 1998 10:55 pm | |
| Mike Smith | Mar 13, 1998 11:10 pm | |
| Bruce Evans | Mar 13, 1998 11:15 pm | |
| Dmitrij Tejblum | Mar 14, 1998 8:23 am | |
| Dmitrij Tejblum | Mar 14, 1998 8:24 am | |
| Dmitrij Tejblum | Mar 14, 1998 8:25 am | |
| Dmitrij Tejblum | Mar 14, 1998 9:59 am | |
| Bruce Evans | Mar 14, 1998 7:08 pm | |
| Dmitrij Tejblum | Mar 15, 1998 2:00 am |
| Subject: | Re: *HEADS UP* Correction to previous postings. | |
|---|---|---|
| From: | Bruce Evans (bd...@zeta.org.au) | |
| Date: | Mar 10, 1998 2:00:07 am | |
| List: | org.freebsd.freebsd-stable | |
Summary of the 3 cases where the whole disk is used for FreeBSD:
Partition table config Slice# passed by bootblocks Slice# to use for "/" ---------------------- --------------------------- ------------------------- Signature != 0xAA55 WHOLE_DISK_SLICE (= 1) COMPATIBILITY_SLICE (= 0) Historical invalid BASE_SLICE + 3 (= 5) COMPATIBILITY_SLICE (= 0) "Valid" (1 big part.) BASE_SLICE + N (0 <= N < 4) BASE_SLICE + N
...
I managed to modify boot2 so I was able to boot with -r1.87 autoconf.c from my dedicated disk (historical invalid). My solution was to leave slice to be WHOLE_DISK_SLICE for a BSD partition starts at sector 0 (this would bloat boot2 by two bytes: a je rel8).
This test would misdetect case 3 as dedicated (which may be best :-) and would also misdetect cases where there are multiple nonempty partitions with the first BSD one starting at sector 0. For complete compatility with the kernel, the boot blocks would have to do a bcmp() of the historical table with the actual table. This takes 0x40 bytes for the table and about 0x20 bytes for the test. I don't think this can be compressed much.
Bruce
To Unsubscribe: send mail to majo...@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message