Re: cgd/suspend

[Jukka Ruohonen <jruohonen%iki.fi@localhost>]

On Fri, Jan 29, 2010 at 10:12:13AM -0600, David Young wrote:
> Ah, that must have happened with cgd's device_t-ification. cgd needs to
> register with PMF when it attaches, and to deregister when it detaches.
> 
> NULL suspend & resume handlers *may* be appropriate for cgd, but let's
> see some justification of that.  Consider whether the driver will do
> something unexpected if you suspend/resume it in isolation, using
> drvctl.

I have the following setup:

        cgd0    /dev/wd0g       (/home)
        cgd1    /dev/wd0b       (swap)

I did:

        swapctl -d /dev/cgd1a
        umount /home

        drvctl -d cgd0
        drvctl -d cgd1

        sysctl -w machdep.sleep_state=3

And everything seemed to be ok.

- Jukka.