On Sun, 26 Nov 2006, Oliver Neukum wrote: > > Second, instead of putting a lock in the buffer_collection you could just > > use the inode's lock. > > That's a layering violation.
Your patch already uses inode->i_mutex. Does that mean it already contains a layering violation? > Plus that lock is used for other things, plus it might be held a pretty long > time (eg the full control timeout) No -- the buffer's lock is used during data transfers. The only time the buffer_collection's lock might overlap a data transfer is during orphan_all_buffers(). Since that happens only when the inode is being removed anyway, I don't think it will hurt to tie up the inode's lock. > > Fourth, you might consider deallocating the sysfs_buffer_collection as > > soon as the last buffer is removed from its list. I don't know if this > > really matters... > > That would mean having to use the inode's lock. As mentioned above, you already use that lock when you create the buffer_collection. I don't see anything wrong with using the lock again when you deallocate it. Alan Stern ------------------------------------------------------------------------- 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 _______________________________________________ linux-usb-devel@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
