Re: Re: Bad: scheduling while atomic!

[Lukas Schrangl <schrangl-hi6Y0CQ0nG0@public.gmane.org>]

Am Donnerstag, 1. April 2004 17:07 schrieb Stefan Seyfried:
> On Thu, Apr 01, 2004 at 08:46:01AM +0200, Lukas Schrangl wrote:
> > Am Mittwoch, 31. März 2004 15:59 schrieb Tomasz Ciolek:
> > >  I get the same error with suspend to disk on a Fujitsu P2120. Ali 1535
> > >  chipset is the common bit...
> > >
> > >  Whats the network card you use?
> >
> > Broadcom BCM 4401
>
> i need to unload bcm4400 before suspend (swsusp, ACPI S4) and reinsert
> after resume, otherwise resume fails. Maybe the same is true for S3.

The problem was enabling both preemptible kernel and MCE. Using only one of 
them everything works, even if b44 is inserted and Local APIC enabled.
Only usb makes problems:
On wakeup I get something like
"Badness in usb/hcd/....c"
and a call trace. Will try to resolve that problem (maybe unloading usbmodules 
works, haven't had time to try yet).
After resume, I am not able to use usb anymore (even unloading/realoading 
doesn't work).
And ACPI- poweroff does not work anymore. When shutting down I can read:
"The system will be haltet immediately.
acpi_power_off called"
or something the like, but no nothing happens; i poweroff manually. On 
"normal" shutdowns (when not having resumed from S4) I don't get the msg 
"acpi_power_off called" and it powers off.
Should I write a bug report because of the preempt/MCE problem and/or the APIC 
error which seems not to have any influence?

Lukas


-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials
Free Linux tutorial presented by Daniel Robbins, President and CEO of
GenToo technologies. Learn everything from fundamentals to system
administration.http://ads.osdn.com/?ad_id\x1470&alloc_id638&op=click