public inbox for linux-acpi@vger.kernel.org
 help / color / mirror / Atom feed
* Reboot on shutdown with USB 2.0
@ 2003-03-28 22:25 Michael Madore
  0 siblings, 0 replies; 3+ messages in thread
From: Michael Madore @ 2003-03-28 22:25 UTC (permalink / raw)
  To: acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f

Hi,

I'm having a strange problem regarding ACPI and USB 2.0 devices 
(External CD-Rom, Jaz drive, etc).  Using 2.4.21-pre5 with the latest 
patch if I issue:

shutdown -h now

The machine shuts off, but half a second later the power comes back on. 
 If I unload the echi driver module, the machine shuts off normally.

Any thoughts on what would cause that kind of behavior?

Thanks!

Mike Madore




-------------------------------------------------------
This SF.net email is sponsored by:
The Definitive IT and Networking Event. Be There!
NetWorld+Interop Las Vegas 2003 -- Register today!
http://ads.sourceforge.net/cgi-bin/redirect.pl?keyn0001en

^ permalink raw reply	[flat|nested] 3+ messages in thread

* RE: Reboot on shutdown with USB 2.0
@ 2003-03-28 22:45 Grover, Andrew
  2003-03-28 23:41 ` Michael Madore
  0 siblings, 1 reply; 3+ messages in thread
From: Grover, Andrew @ 2003-03-28 22:45 UTC (permalink / raw)
  To: Michael Madore, acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f

> From: Michael Madore [mailto:mmadore-JChM1cqIHx8AvxtiuMwx3w@public.gmane.org] 

> I'm having a strange problem regarding ACPI and USB 2.0 devices 
> (External CD-Rom, Jaz drive, etc).  Using 2.4.21-pre5 with the latest 
> patch if I issue:
> 
> shutdown -h now
> 
> The machine shuts off, but half a second later the power 
> comes back on. 
>  If I unload the echi driver module, the machine shuts off normally.
> 
> Any thoughts on what would cause that kind of behavior?

Is this an SMP system? Just curious.

What we should be doing (but aren't) is telling each device's driver on
the system to deactivate its device before we turn off the system power.
This gives the driver the chance to disable that device's interrupt
generation. We haven't gotten that far in the driver model work, so what
we are doing is disabling interrupts and then powering down. This mostly
works, but isn't really how we want things to work, and is pretty much
doomed to fail on SMP systems.

I don't know why the ehci device would specifically be the one to cause
problems...

Regards -- Andy


-------------------------------------------------------
This SF.net email is sponsored by:
The Definitive IT and Networking Event. Be There!
NetWorld+Interop Las Vegas 2003 -- Register today!
http://ads.sourceforge.net/cgi-bin/redirect.pl?keyn0001en

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Reboot on shutdown with USB 2.0
  2003-03-28 22:45 Reboot on shutdown with USB 2.0 Grover, Andrew
@ 2003-03-28 23:41 ` Michael Madore
  0 siblings, 0 replies; 3+ messages in thread
From: Michael Madore @ 2003-03-28 23:41 UTC (permalink / raw)
  To: Grover, Andrew; +Cc: acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f

Grover, Andrew wrote:

>>From: Michael Madore [mailto:mmadore-JChM1cqIHx8AvxtiuMwx3w@public.gmane.org] 
>>    
>>
>
>  
>
>>I'm having a strange problem regarding ACPI and USB 2.0 devices 
>>(External CD-Rom, Jaz drive, etc).  Using 2.4.21-pre5 with the latest 
>>patch if I issue:
>>
>>shutdown -h now
>>
>>The machine shuts off, but half a second later the power 
>>comes back on. 
>> If I unload the echi driver module, the machine shuts off normally.
>>
>>Any thoughts on what would cause that kind of behavior?
>>    
>>
>
>Is this an SMP system? Just curious.
>  
>
Yes, this particular system is SMP.  I'll have to try it with a UP 
kernel just to see if the problem persists.

>What we should be doing (but aren't) is telling each device's driver on
>the system to deactivate its device before we turn off the system power.
>This gives the driver the chance to disable that device's interrupt
>generation. We haven't gotten that far in the driver model work, so what
>we are doing is disabling interrupts and then powering down. This mostly
>works, but isn't really how we want things to work, and is pretty much
>doomed to fail on SMP systems.
>
>I don't know why the ehci device would specifically be the one to cause
>problems...
>  
>
That makes sense, thanks for the explaination.  As a workaround, perhaps 
we can modify the init scripts to unload the ehci driver during shutdown.

Mike Madore



-------------------------------------------------------
This SF.net email is sponsored by:
The Definitive IT and Networking Event. Be There!
NetWorld+Interop Las Vegas 2003 -- Register today!
http://ads.sourceforge.net/cgi-bin/redirect.pl?keyn0001en

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2003-03-28 23:41 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2003-03-28 22:45 Reboot on shutdown with USB 2.0 Grover, Andrew
2003-03-28 23:41 ` Michael Madore
  -- strict thread matches above, loose matches on Subject: below --
2003-03-28 22:25 Michael Madore

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox