Re: [PATCH 4/4] USB: add urb->unlinked field

On Tue, Aug 21, 2007 at 05:51:22PM -0400, Alan Stern wrote:
> On Tue, 21 Aug 2007, Pete Zaitcev wrote:
> 
> > On Tue, 21 Aug 2007 13:53:38 -0700, Greg KH <[email protected]> wrote:
> > 
> > > > > Hm, is this field now safe to be used by drivers to determine if the urb
> > > > > is still being handled by the host controllers also?
> > > > 
> > > > You mean is the urb->status conversion finished?  Not yet.  I've got 
> > > > more patches still to test and submit.  It's getting there...
> > > 
> > > No, I mean is this something that drivers can use to check if the urb is
> > > finished with the HCD, like they currently check for -EINPROGRESS today.
> > 
> > Drivers must maintain this status by setting and clearing a flag.
> > They set it before submitting, clear it in callback or if submission
> > fails. I thought we had your agreement on this as a stack maintainer.
> > In my practice so far it worked perfectly. Is there a specific driver
> > you have in mind where this requirement is inconvenient?
> 
> I think Greg is worried that some driver might try using the new field 
> in a bogus manner.  There was never any intention to change the current 
> requirement.

Yes, that is exactly what I was asking about.

And yes, Pete, I still want to see the driver set their own flag for urb
usage (hm, perhaps we should just make a default one in the urb itself
for this...) and not abuse this new field.

Alan, thanks for clearing this up.

greg k-h

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
[email protected]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel