From mboxrd@z Thu Jan  1 00:00:00 1970
Message-ID: <eedc56c203579abe7a88a48d6d4f3079@plan9.bell-labs.com>
From: "Russ Cox" <rsc@plan9.bell-labs.com>
To: 9fans@cse.psu.edu
Subject: Re: [9fans] exception/interrupt 14 bug
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
Date: Mon, 13 May 2002 21:18:56 -0400
Topicbox-Message-UUID: 906b09e6-eaca-11e9-9e20-41e7f4b1d025

> The fix doesn't seem to be related to my problem though -
> exception/interrupt 14 happens much earlier in my case, before the kernel is
> fully loaded. The new floppy still panics.

It was the same problem, though.  Here's your register
dump:

apm ax=f000 cx=f000 dx=0 di=1200 ebx=57a0 esi=c03e6
using fd0!dos!plan9.ini
dev A0 port 1F0 config 0C5A capabilities 2F00 mwdma 0007 udma 043F
pcirouting: 8086/7112 at pin 4 irq 7
found 9PCFLOP .GZ attr 0x0 start 0xf0 len 1168843
.....FLAGS=10246 TRAP=e ECODE=0 PC=80015b87
AX a48003b7 BX 8003b784 CX 00000800 DX 00000000
SI 8002dcc0 DI 8003b7a4 BP 8003b780
CS 0010 DS 0008 ES 0008 FS 0008 CS 0008
CR0 80000011 CR2 a48003b9 CR3 0000c000
panic: exception/interrupt 14

57a0+03e6 is 5b86.  Toss in the segment register and
you've got 80015b86.  Note that you died at 80015b87.

bootpass+0x5fc 0x80015b84	MOVL	0x24(BP),AX
bootpass+0x5ff 0x80015b87	MOVBSX	0x2(AX),AX

The magic increment turned the first instruction
into

bootpass+0x5fc 0x80015b84	MOVL	0x25(BP),AX

and you died dereferencing the shifted value.

As for why the disk didn't work, well, let's just
say that caches are not my friend this week.

If you download a disk now, it should work.

Russ