Re: [PATCH 4/4] USB: export autosuspend delay in sysfs

Greg KH <[email protected]>
Newsgroups gmane.linux.usb.devel
Message-ID <[email protected]>
On Fri, Feb 23, 2007 at 06:23:08PM -0800, David Brownell wrote:
> On Friday 23 February 2007 5:05 pm, Alan Stern wrote:
> > On Fri, 23 Feb 2007, Greg KH wrote:
> 
> > > I have heard rumors that both Intel and mvista are starting to dig up
> > > the old patches and want to get this stuff all sorted out and into the
> > > main kernel tree.  Hopefully something comes of it...
> 
> Certainly I'd anticipate much more email.  Previous discussions have been
> inconclusive, IMO basically because they were too much in the category
> of "solution in search of problem".  Also, ISTR they didn't even start
> to get near device-specific power management... but, maybe you're not
> thinking about the patches I'm recalling!

No, I'm thinking of the same mess :)

And yes, I would expect a lot of email about this too, or at least I
sure hope so...

> > My crystal ball is looking particularly cloudy today.  Eventually 
> > something will have to be done, but will it be done independently by each 
> > subsystem?  Or will there be a centralized PM core API with a 
> > subsystem-specific hook?  I have no idea.
> 
> If I were to bet, I'd say per-subsystem approaches will work better
> than any Grand Unified/Centralized Scheme.  Per-driver approaches will
> also be needed.

Agreed.

> > So it comes down to the same question: Should I submit this USB-only 
> > facility for inclusion now, with the understanding that it may need to be 
> > changed later?
> 
> I'd say submit it, but with a clear statement about how changeable the
> interface is.  Such a statement would be a kind of new ground for Linux,
> which is needed ... some folk take the wrong lesson about stability of
> kernel interfaces, and want all interfaces to be stable.  That's not a
> realistic expectation (IMO) in cases where some experimentation is needed
> to reach the eventual answer.  But it _is_ realistic to expect that if
> an interface isn't fully cooked, it will be declared as such ... so that
> userspace developers will at least know they need to communicate with the
> kernel developers if they plan to make heavy use of that interface.

That's what Documentation/ABI/ is there for :)

thanks,

greg k-h

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
[email protected]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.