lkml.org 
[lkml]   [1998]   [Jan]   [12]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Not rebooting.
On Mon, 12 Jan 1998, Pavel Machek wrote:

> Hi!
>
> > Specifying 'reboot=bios' does work however, which leads me to a question:
> >
> > If the default method of rebooting without the BIOS (using keyboard/triple
> > fault) doesn't work, why does the current code sit in an infinite loop when
> > it would be just as easy to attempt that a certain number of times and then
> > attempt the BIOS instead?
>
> Because triplefaulting means crashing processor. And you can not do
> anything after you crashed you CPU. And changing methods will not
> work, IMO: BIOS simply will not reboot, and it may not even return control.
>
Note that the 'keyboard' controller is really a big port with a built-in
CPU! One of the things it can do is hit the reset switch (really). When
attempting to reboot using the keyboard controller, it actually activates
the CPU reset pin --if the motherboard is not broken by design.

Now, when the CPU is in real mode, power-on or reset makes it jump to
FFFF:0000. This it the 'reset' vector. With '386s and higher, the restart
code is made visible at 0xfffffff0 in linear address space after reset,
which should allow a restart directly from protected mode if the BIOS
properly sets up 16-bit segments, etc. Both of my Pentiums do this, but my
'486 does not. Therefore, this capability is not useful.

What all this means is that it is difficult to reset the computer if the
reset switch doesn't work.


Cheers,
Dick Johnson
***** FILE SYSTEM MODIFIED *****
Penguin : Linux version 2.1.70 on an i586 machine (66.15 BogoMips).
Warning : It's hard to remain at the trailing edge of technology.


\
 
 \ /
  Last update: 2005-03-22 13:41    [W:0.034 / U:1.952 seconds]
©2003-2017 Jasper Spaans. hosted at Digital OceanAdvertise on this site