uevent / call_usermoderrelated breakage on suspend to disk [Was: Re: PCMCIA-related breakage on suspend to disk]

[Russell King]

On Wed, Dec 07, 2005 at 08:26:30PM +0100, Pavel Machek wrote:
> Hi!
> 
> > What to do?
> > a) fix send_uevent or call_usermodehelper to not lock up in the
> > 	resume-when-suspending-to-disk-path
> 
> This is the right way to go I guess.
> 
> > b) make device_add fail if in
> > 	resume-when-suspending-to-disk-path
> 
> That may have disable suspend-to-USB-disk and similar weird
> scenarios. Not sure if that's a big loss.
> 
> > c) disable device-adding in drivers/pcmcia/cs.c:socket_resume() if in
> > 	resume-when-suspending-to-disk-path
> 
> Probably simplest... 

No.  The question is - why does this work on previous kernels and why
has it only just apparantly broken?

I use this facility all the time, albiet on the kernel version it was
introduced with.  And no, because of the crap surrounding the PCMCIA
ioctl, sorry I can't test later kernels.

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:  2.6 Serial core