Re: APM oops in 2.1.127

Alan Olsen (alano@adams.pcx.ncd.com)
Fri, 20 Nov 1998 10:00:54 -0800


On Nov 20, 2:22am, Stephen.Rothwell@canb.auug.org.au wrote:
> Subject: Re: APM oops in 2.1.127
> Hi,
>
> Thomas Molina <tmolina@probe.net> writes:
> > System is RedHat 5.2, kernel 2.1.127. This is a new motherboard with
> > apm support so I compile it in. The support is recognized based on boot
> > up messages. The motherboard is a TMC TIVG+ with an MVP3 (82C598AT)
> > chipset. Processor is an AMD K6-2 at 350MHz. I modified the shutdown
> > scripts to call halt -p for powerdown after halt. After the "system
> > halted" message I get a "Power down." message after which is a general
> > protection fault. Since I'm unsure how to set up the system so it
> > captures the debugging info after this, I've copied some of it by hand.
> > I can do a line-by-line copy if it will help run down a debugging
> > process:
> >
> > General Protection Fault: f000
> > EIP 0050: [<00008abb>]
> >
> > Call trace shows (with hand-correlated symbols from the appropriate
> > system map:
> >
> > c010632d (apm_set_power_state)
> > c0106370 (apm_power_off)
> > c0107135 (machine_power_off)
> > c0112b5b (sys_reboot)
> > c01c325a (tvecs)
> > c010e6e5 (send_sig_info)
> > c010e964 (kill_something_info)
> >
> > Then I get the following message:
> > Unable to handle kernel paging request at virtual address 00008abb
> > current->tss.cr3=0193100 %cr3=0193100 (tty_write?)
> >
> > then I get an oops 0000
> > EIP 0010: [<0108c43>] (show_registers?)
> > with most of the same call trace info as before
> >
> > Suggestions anyone?
>
> You have a broken BIOS! For some reason it is referencing (real mode)
> segment f000. It obviously expect to be in real mode (as opposed to
> protected mode) when asked to power down. Try talking to your
> supplier or either you motherboard, or BIOS.

Will bios shadowing reference real mode at f000? (I turn bios shadowing off,
but that might explain what he is seeing.)

-- 
Alan Olsen

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/