On Wed, Sep 03, 2003 at 09:06:36PM +0200, Oliver Neukum wrote: > As far as I can tell, the problem is that the reference count is not per > device, but per struct file. Thus you can get several calls to release() > for a single device. The question that needs to be answered is whether > the number of times release() is called is equal to the number of times > open() is called.
If I may add a vote, unbalanced opens and releases are a bad idea. There used to be a SunOS where it was the case over a fork, it was insane. Essentially every driver resorted to exclusion (e.g. only one open allowed), because it was the only way to keep softcarriers consistent. If our framework allows unablanced releases, it should be fixed in the framework, not in drivers. -- Pete ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
