From: "David N. Welton" <davidw@eidetix.com>
To: Sascha Wilde <wilde@sha-bang.de>
Cc: Dmitry Torokhov <dtor_core@ameritech.net>,
LKML <linux-kernel@vger.kernel.org>,
Vojtech Pavlik <vojtech@suse.cz>
Subject: Re: 2.6 kernel won't reboot on AMD system - 8042 problem?
Date: Thu, 12 Aug 2004 19:00:04 +0200 [thread overview]
Message-ID: <411BA214.2060306@eidetix.com> (raw)
In-Reply-To: <20040811175613.GA829@kenny.sha-bang.local>
Sascha Wilde wrote:
>>> Is PS/2 supposed to support hotpluging at all? I guess it's not,
>>> but I may be wrong...
>>
>> Yes it is, at least with newer (or rather not ancient) hardware...
>
>
> well, so the patch obviously can't be a final solution...
Yes, it doesn't strike me as being ideal.
More playing with the triple fault leads me to this:
static int i8042_command(unsigned char *param, int command)
{
unsigned long flags;
int retval = 0, i = 0;
spin_lock_irqsave(&i8042_lock, flags);
retval = i8042_wait_write();
if (!retval) {
dbg("%02x -> i8042 (command)", command & 0xff);
i8042_write_command(command & 0xff);
}
=====> A
if (!retval)
for (i = 0; i < ((command >> 12) & 0xf); i++) {
if ((retval = i8042_wait_write())) break;
dbg("%02x -> i8042 (parameter)", param[i]);
i8042_write_data(param[i]);
}
=====> B
Point 'B' is the "point of no reboot", at A we can still triple fault it
and get a nice normal reboot, so it's the parameter that it's writing
that it causing it to go funny.
I noticed something odd though... I reported this in another email:
*With keyboard* :
mice: PS/2 mouse device common for all mice
drivers/input/serio/i8042.c: 20 -> i8042 (command) [0]
drivers/input/serio/i8042.c: 65 <- i8042 (return) [0]
*Without keyboard* :
mice: PS/2 mouse device common for all mice
drivers/input/serio/i8042.c: fa <- i8042 (flush, aux) [0]
drivers/input/serio/i8042.c: 20 -> i8042 (command) [0]
drivers/input/serio/i8042.c: 9a <- i8042 (return) [0]
I noticed that
0x9a is the 'inverse' of 0x65. If I set it to 0x65 and write that out,
it still reboots afterwards! Hrm. I don't know what that means
exactly, but apparently it *is* possible to write something to the
controller and have it keep going.
Sascha, if you want to test it out, try this in i8042_controller_init,
at about line 724 (near this: i8042_initial_ctr = i8042_ctr;)
{
unsigned char pram;
pram = (~i8042_ctr) & 0xff ;
i8042_command(&pram, I8042_CMD_CTL_WCTR);
}
{
static struct
{
unsigned short size __attribute__ ((packed));
unsigned long long * base __attribute__ ((packed));
} no_idt = { 0, 0 };
__asm__ __volatile__("lidt %0": :"m" (no_idt));
__asm__ __volatile__("int3");
}
It reboots!
--
David N. Welton
davidw@eidetix.com
next prev parent reply other threads:[~2004-08-12 17:00 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-08-11 14:14 2.6 kernel won't reboot on AMD system - 8042 problem? Dmitry Torokhov
2004-08-11 17:56 ` Sascha Wilde
2004-08-12 17:00 ` David N. Welton [this message]
2004-08-12 17:23 ` David N. Welton
2004-08-13 21:29 ` Sascha Wilde
2004-08-12 20:13 ` Vojtech Pavlik
2004-08-13 10:13 ` David N. Welton
2004-08-13 12:03 ` Vojtech Pavlik
2004-08-13 12:58 ` David N. Welton
[not found] <auto-000000462036@appliedminds.com>
2004-08-09 8:28 ` David N. Welton
2004-08-10 9:37 ` Sascha Wilde
2004-08-10 15:38 ` James Lamanna
[not found] <4112A626.1000706@appliedminds.com>
2004-08-06 8:22 ` David N. Welton
2004-08-06 16:55 ` James Lamanna
2004-08-08 12:18 ` Sascha Wilde
2004-08-08 15:05 ` Dmitry Torokhov
2004-08-11 20:06 ` Sascha Wilde
-- strict thread matches above, loose matches on Subject: below --
2004-07-28 17:51 2.6 kernel won't reboot on AMD system (no, not the BIOS...) David N. Welton
2004-08-05 12:48 ` 2.6 kernel won't reboot on AMD system - 8042 problem? David N. Welton
2004-08-05 19:25 ` Sascha Wilde
2004-08-11 6:31 ` Dmitry Torokhov
2004-08-11 8:36 ` David N. Welton
2004-08-11 12:27 ` Vojtech Pavlik
2004-08-11 12:45 ` David N. Welton
2004-08-11 13:43 ` Sascha Wilde
2004-08-11 14:17 ` Vojtech Pavlik
2004-08-11 13:55 ` David Ford
2004-08-11 20:14 ` Sascha Wilde
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=411BA214.2060306@eidetix.com \
--to=davidw@eidetix.com \
--cc=dtor_core@ameritech.net \
--cc=linux-kernel@vger.kernel.org \
--cc=vojtech@suse.cz \
--cc=wilde@sha-bang.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox