On 20 Jun 02 at 20:49, [email protected] wrote:
>
> Now that you tell me that these things are set at initialization
> and never changed, the initialization must be wrong. And indeed,
> it says "__devexit_p(uhci_stop)" and this yields NULL.
Now when kernel tries to shutdown devices on poweroff, should
not we change __devexit_p() meaning? Like always build kernel with
hotplug enabled?
Thanks,
Petr Vandrovec
[email protected]
On Thu, Jun 20, 2002 at 09:40:27PM +0200, Petr Vandrovec wrote:
> On 20 Jun 02 at 20:49, [email protected] wrote:
> >
> > Now that you tell me that these things are set at initialization
> > and never changed, the initialization must be wrong. And indeed,
> > it says "__devexit_p(uhci_stop)" and this yields NULL.
>
> Now when kernel tries to shutdown devices on poweroff, should
> not we change __devexit_p() meaning? Like always build kernel with
> hotplug enabled?
You should always enable CONFIG_HOTPLUG anyway :)
Seriously, you are correct, we will probably have to go back and audit
the __devexit functions when we put poweroff support in.
thanks,
greg k-h