Hello,
I have a strange Problem with all 2.6.x kernels (and only with
these): my box refuses to reboot.
The Problem:
All versions of Linux 2.6.x (I started testing with 2.6.test6 and the
latest version I tied was 2.6.4 release) cause my System to hang when
I try to reboot (or power off).
The Hardware:
The system in question is a x86 PC with:
AMD Athlon(tm) Processor stepping 02 CPU (classic 800MHz)
AMD Irongate chipset
This list is quite short, for I have no idea which components could be
related to the problem. I will provide you with any further
information needed or send you a full kernel startup log if wanted.
What I tried:
The problem occurs at the end of any shut-down process, right when
the system should reboot (or power off). I tried different Kernel
versions (form 3.6.test6 to 2.6.4) using different compiler
tool-chains and configurations. It also tried a kernel 2.6.1 from a
"Knoppix" CD distribution, so this is unrelated to my compiler.
I tried booting into single user and rebooting via sysreq-keys, it
made the machine hang, too -- so I would say the userland is not
involved.
During digging in the dark I also tried different power-management
configurations with APM and ACPI, no chance (normally I use APM).
This is only a 2.6.x matter, all earlier Kernels I used (including
some 2.2.x versions and nearly every 2.4.x version), the netbsd
kernel (1.6.1 and 1.6.2) and the grub bootloader are able to reboot
the system as expected.
This problem is 100% reproducible on my box, but I haven't yet found
any other system showing this symptoms.
Sorry for this very vague report, but I have simply no idea, what
might be the cause.
Please CC me in any answers, I'm not subscribed to the lkml.
cheers
sascha
ps. please don't mind my worse English.
--
Sascha Wilde
... mein Opa [...] w?rde an dieser Stelle zu Dir sagen: Junge, such Dir
ne Frau, bau Dir ein Haus, mach ein Kind und la?' die Finger von dem Zeug,
das Du gerade machst. -- Michael Winklhofer in d.a.e.auktionshaeuser
Curious that this happens
1. only on 2.6
2. with acpi=off and "acpi=on"
Would be good to know that the system is running the latest BIOS.
cheers,
-Len
On Fri, 2004-03-12 at 18:36, Sascha Wilde wrote:
> Hello,
>
> I have a strange Problem with all 2.6.x kernels (and only with
> these): my box refuses to reboot.
>
>
> The Problem:
>
> All versions of Linux 2.6.x (I started testing with 2.6.test6 and the
> latest version I tied was 2.6.4 release) cause my System to hang when
> I try to reboot (or power off).
>
>
> The Hardware:
>
> The system in question is a x86 PC with:
> AMD Athlon(tm) Processor stepping 02 CPU (classic 800MHz)
> AMD Irongate chipset
>
> This list is quite short, for I have no idea which components could be
> related to the problem. I will provide you with any further
> information needed or send you a full kernel startup log if wanted.
>
>
> What I tried:
>
> The problem occurs at the end of any shut-down process, right when
> the system should reboot (or power off). I tried different Kernel
> versions (form 3.6.test6 to 2.6.4) using different compiler
> tool-chains and configurations. It also tried a kernel 2.6.1 from a
> "Knoppix" CD distribution, so this is unrelated to my compiler.
>
> I tried booting into single user and rebooting via sysreq-keys, it
> made the machine hang, too -- so I would say the userland is not
> involved.
>
> During digging in the dark I also tried different power-management
> configurations with APM and ACPI, no chance (normally I use APM).
>
> This is only a 2.6.x matter, all earlier Kernels I used (including
> some 2.2.x versions and nearly every 2.4.x version), the netbsd
> kernel (1.6.1 and 1.6.2) and the grub bootloader are able to reboot
> the system as expected.
>
>
> This problem is 100% reproducible on my box, but I haven't yet found
> any other system showing this symptoms.
>
> Sorry for this very vague report, but I have simply no idea, what
> might be the cause.
>
> Please CC me in any answers, I'm not subscribed to the lkml.
>
> cheers
> sascha
>
> ps. please don't mind my worse English.
> --
> Sascha Wilde
> ... mein Opa [...] w?rde an dieser Stelle zu Dir sagen: Junge, such
> Dir
> ne Frau, bau Dir ein Haus, mach ein Kind und la?' die Finger von dem
> Zeug,
> das Du gerade machst. -- Michael Winklhofer in d.a.e.auktionshaeuser
> -
> To unsubscribe from this list: send the line "unsubscribe
> linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
On Sun, Mar 14, 2004 at 01:08:25AM -0500, Len Brown wrote:
> Curious that this happens
> 1. only on 2.6
yes, but true: any other reboot code I've come across (including
those of the other Linux kernel versions) work as expected on the box.
> 2. with acpi=off and "acpi=on"
yust retested it with the latest (2.6.4) kernel -- yes, regardless of
which (or none) powermanagement is used, the reboot freezes the mashine.
> Would be good to know that the system is running the latest BIOS.
afaik I'm using the latest available BIOS for the board.
btw. the toppic is false, that should be
"Kernel 2.6.x reboot hangs on AMD Athlon System"
cheers
sascha
--
Sascha Wilde
"Gimme about 10 seconds to think for a minute..."
On Sun, Mar 14, 2004 at 07:35:58PM +0100, Sascha Wilde wrote:
> On Sun, Mar 14, 2004 at 01:08:25AM -0500, Len Brown wrote:
> > Curious that this happens
> > 1. only on 2.6
>
> yes, but true: any other reboot code I've come across (including
> those of the other Linux kernel versions) work as expected on the box.
>
> > 2. with acpi=off and "acpi=on"
>
also try with noapic boot option. It was supposedly fixed (cause a
problem with my system) but may still causing problems for you.
> yust retested it with the latest (2.6.4) kernel -- yes, regardless of
> which (or none) powermanagement is used, the reboot freezes the mashine.
>
> > Would be good to know that the system is running the latest BIOS.
>
> afaik I'm using the latest available BIOS for the board.
>
> btw. the toppic is false, that should be
> "Kernel 2.6.x reboot hangs on AMD Athlon System"
>
> cheers
> sascha
> --
> Sascha Wilde
> "Gimme about 10 seconds to think for a minute..."
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
Micha Feigin wrote:
> On Sun, Mar 14, 2004 at 07:35:58PM +0100, Sascha Wilde wrote:
> > On Sun, Mar 14, 2004 at 01:08:25AM -0500, Len Brown wrote:
> > > Curious that this happens
> > > 1. only on 2.6
> >
> > yes, but true: any other reboot code I've come across (including
> > those of the other Linux kernel versions) work as expected on the
> > box.
> >
> > > 2. with acpi=off and "acpi=on"
>
> also try with noapic boot option. It was supposedly fixed (cause a
> problem with my system) but may still causing problems for you.
Just tried that, no effect... :-(
Something must be strangely different between the way 2.6.x tries to
reboot the System and any other Kernel-version does. Who is the
Maintainer of this code? Or maybe even better, who did the changes,
made to the reboot-code in 2.6? (There is a ne file reboot.c, but
without any author notice)
Please cc me in any reply, I'm not subscribed to the lkml.
cheers
--
Sascha Wilde : "I heard that if you play the Windows CD backward, you
: get a satanic message. But that's nothing compared to
: when you play it forward: It installs Windows...."
: -- G. R. Gaudreau