Re: [RFC] usbfs: Add ioctls for runtime suspend and resume

[Greg KH <greg@kroah.com>]

On Thu, Jul 25, 2019 at 10:10:27AM +0100, Mayuresh Kulkarni wrote:
> On Fri, 2019-07-05 at 14:51 -0400, Alan Stern wrote:
> > On Wed, 3 Jul 2019, Mayuresh Kulkarni wrote:
> > 
> > > 
> > > As you had mentioned in one of the comment before, the only addition
> > > to
> > > the patch I have locally is -
> > > usbfs_notify_resume() has usbfs_mutex lock around list traversal.
> > > 
> > > Could you please send the patch for review? Please note, I think I
> > > am
> > > not a part of linux-usb mailing-list, so probably need to be in cc
> > > to
> > > get the patch email. Do let me know if something else is needed from
> > > me.
> > Here it is.  There are two changes from the previous version:
> > 
> > 1.	This is rebased on top of a separate patch which Greg has 
> > 	already accepted:
> > https://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb.git/commit?
> > id=ffed60971f3d95923b99ea970862c6ab6a22c20f
> > 
> > 2.	I implemented Oliver's recommendation that the
> > WAIT_FOR_RESUME
> > 	ioctl should automatically do FORBID_SUSPEND before it returns, 
> > 	if the return code is 0 (that is, it wasn't interrupted by a 
> > 	signal).
> > 
> > Still to do: Write up the documentation.  In fact, the existing
> > description of usbfs in Documentation/driver-api/usb/usb.rst is sadly
> > out of date.  And it deserves to be split out into a separate file of
> > its own -- but I'm not sure where it really belongs, considering that
> > it is an API for userspace, not an internal kernel API.
> > 
> > Greg, suggestions?
> 
> Hi Greg,
> 
> Did you got a chance to look into the above documentation query by Alan?
> How should we go about documenting these new IOCTLs?

Not yet, sorry, dealing with the backlog of patches after the merge
window closed.

Give me a week or so...

But if you want to try your hand at it first, it's always easier to
review a patch than it is to come up with a new one.

thanks,

greg k-h