lkml.org 
[lkml]   [2008]   [Feb]   [6]   [last100]   RSS Feed
Views: [more markup]  [less markup]  [headers]  [forward] 
 
Messages in this thread
/
DateWed, 6 Feb 2008 15:23:49 -0800
FromGreg KH <>
SubjectRe: [PATCH 2/2] - SH/Dreamcast - fix maple bus bugs
On Wed, Feb 06, 2008 at 11:05:14PM +0000, Adrian McMenamin wrote:> On Wed, 2008-02-06 at 15:01 -0800, Greg KH wrote:> > On Wed, Feb 06, 2008 at 10:53:51PM +0000, Adrian McMenamin wrote:
> > > -	dev->function = function;> > > -	dev->dev.bus = &maple_bus_type;> > > -	dev->dev.parent = &maple_bus;> > > -	dev->dev.release = &maple_release_device;> > > -	retval = device_register(&dev->dev);> > > -	if (retval) {> > > -		printk(KERN_INFO> > > -		       "Maple bus: Attempt to register device (%x, %x) failed.\n",
> > > -		       dev->port, dev->unit);> > > -		maple_free_dev(dev);> > > +	mdev->function = function;> > > +	mdev->dev.release = &maple_release_device;
> > > +	if (mdev->registered == 0) {> > > +		retval = device_register(&mdev->dev);> > > +		if (retval) {> > > +			printk(KERN_INFO> > > +			"Maple bus: Attempt to register device"
> > > +			" (%x, %x) failed.\n",
> > > +			mdev->port, mdev->unit);
> > > +			maple_free_dev(mdev);
> > > +			mdev = NULL;
> > > +			return;
> > > +		}
> > > +		mdev->registered = 1;
> > >  	}
> > > -	dev->registered = 1;> > > > I think you are still papering over the real problem here of trying to
> > register a device twice.  That's something you should never be doing,
> > and your bus walking logic must be incorrect...> > > I just left that in as a sanity check. It should never be acted on. I
> can remove it if you prefer.
Please do, that way, and real problems will show up with that warning
you were getting that started all of this work :)

thanks,

greg k-h


\
 
 \ /
  Last update: 2008-02-07 00:29    [from the cache]
©2003-2008