Re: Re: 2.6.0-test8 mad clock rate drifts and sleeping function ...

[john stultz <johnstul@us.ibm.com>]

On Fri, 2003-10-24 at 04:06, Roland Lezuo wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> > So you're seeing time run twice as fast overall? Are you running with
> > NTP?  Do you have any sort of hardware power management on the system?
> > Do you have any more details about the system?
> 
> I used to run NTP but I thought I didn't do that when I epxerienced that clock 
> drift. The Box (after a restart) is up for 1 day and 13 hours now, there is 
> no clock drift any more...
> ...the xmms problem disappeared by recompileing the xmms-alsa plugin (which is 
> strange for me), but it is gone...
> 
> 
> the only thing left:
> 
> Debug: sleeping function called from invalid context at 
> include/asm/uaccess.h:473
> in_atomic():0, irqs_disabled():1
> Call Trace:
>  [<c0120150>] __might_sleep+0xa0/0xd0
>  [<c010d40a>] save_v86_state+0x6a/0x200
>  [<c010ca67>] do_IRQ+0x117/0x160
>  [<c010a49e>] work_notifysig_v86+0x6/0x14
>  [<c010a44b>] syscall_call+0x7/0xb
> 
> ... there are more of this called from different source files...

Dunno about that one, but please let me know if you later find you can
reproduce the time drift issue. 

thanks
-john