On Sat, Dec 15, 2007 at 08:41:24PM -0800, Jeff Rizzo wrote:
> It seems to me that drvctl(8) is pretty useful - and it requires the
> drvctl pseudo-device in the kernel.  It's necessary for those trying to
> use GPT on their disks, for example.  It seems to me it should be in
> GENERIC on most ports - though I certainly wouldn't undertake to blindly
> enable it on ports I don't use.
> 
> What do other folks think?  At the very least, I'd think on i386 and
> amd64 - I could test it on macppc as well, though I haven't (yet) had
> need of it there.

Jeff,

I believe drvctl should be present by default, too.

BTW, I have been using drvctl to detach/re-attach some devices on a
Soekris net4521.  I'm finding that many devices have broken or incomplete
detachment routines.

I have added to drvctl some options for suspending sub-trees of the
device tree, and for resuming either sub-trees or individual devices.
This is handy both for testing, and for exploiting suspend/resume on
systems without ACPI.

Dave

-- 
David Young             OJC Technologies
dyoung@ojctech.com      Urbana, IL * (217) 278-3933 ext 24