Date: Wed, 06 Apr 2005 18:26:07 +0200 From: Phil Schulz <ph.schulz@gmx.de> To: Phil Schulz <ph.schulz@gmx.de>, freebsd-acpi@freebsd.org Subject: Re: power off issues (introduced btw 05/28/04 and 05/29/04) Message-ID: <42540D9F.9070504@gmx.de> In-Reply-To: <4252AE55.4090706@gmx.de> References: <4252AE55.4090706@gmx.de>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi again!
I tried to dig deeper and found out that the system started to behave
incorrectly after the default of acpi_disable_on_poweroff was changed
from 1 to 0 on 05/28/2004 at around 7:17 GMT. So I thought a solution
might be to change it back to 1 but it seems that not only the sysctl
but also the infrastructure behind it is gone in more recent versions of
-stable.
It seems that the system does OK until the call to
AcpiEnterSleepState() in acpi_shutdown_final(), line 1611 of
sys/dev/acpica/acpi.c Rev. 1.186.2.12 and then goes crazy.
Going further with the infamous add-a-lot-of-printfs method I traced
it until the call to AcpiHwRegisterWrite (ACPI_MTX_DO_NOT_LOCK,
ACPI_REGISTER_PM1A_CONTROL, PM1AControl) in AcpiEnterSleepState(),
line 423 of sys/contrib/dev/acpica/hwsleep.c Rev. 70
At that point, PM1AControl is 0x3c01
I haven't bothered to look further since I don't really know what to
look for. Is there anybody out there who could give me a hint? Does it
even make sense to keep digging or is the hardware so horribly broken
that there is almost no chance to fix it?
The whole debugging process is kind of painfull since the stupid
laptop does not have a serial port and the firewire device is already
shut off at that point of the shutdown procedure so I don't have any
debugger :-(
Regards,
Phil.
On 04/05/05 17:27, Phil Schulz wrote:
> Hi!
>
> My laptop does not shut down properly when I issue "halt -p",
> "shutdown -p now" or hit the power button. When it is supposed to be
> shut off, the fan spins up and lots of random colored dots appear all
> over the display. The problem was introduced between May 28th, 2004 and
> May 29th, 2004.
> njl@ mentions that he has finnished a some work on ACPI events possibly
> on May 28th in a message [1] to this list, so I think it might be
> related to my problem.
> I know it is kind of late to report this but I ran RELENG_5_2 at the
> time and did not notice anything until upgrading to RELENG_5 after 5.3
> was released.
> The laptop is a Fujitsu-Siemens Amilo M-7400 using the latest BIOS
> version provided by Fujitsu-Siemens. It runs FreeBSD 5-Stable as of Apr
> 05, 2005 around 13:00 GMT. The other information as requested by the
> handbook in section 11.16 is available at
>
> http://www.home.fh-karlsruhe.de/~scph0020/downloads/freebsd/acpi/
>
> Can anybody help me? I'd be happy to provide more information on
> request and I can also provide console access for anybody who is
> interested.
>
> Kind regards,
>
> Phil.
>
> P.S.: Please cc: me on replies since I am not on the list.
>
> [1] http://lists.freebsd.org/pipermail/freebsd-acpi/2004-May/000196.html
>
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?42540D9F.9070504>