On Fri, Oct 21, 2005 at 08:39:09AM -0700, Pete Zaitcev wrote: > On Wed, 19 Oct 2005 11:48:57 -0700, Greg KH <[EMAIL PROTECTED]> wrote: > > > > Also, do you see that rwlock which IS TAKEN FOR WRITE ONLY? > > > Why? The only explanation is: the author intended read locks > > > in notifier_call_chain, but either forgot them, or they were > > > dropped by other person. > > > Ick, you are right. I'll go redo my notifier code to get this right. > > Greg, here's some info just in case it saves you some effort if you > decide to work inside the notifier. I did some research and composed > a message to you, but it disappeared somehow, and then I went away > for two days... Apparently, someone replaced spinlocks with rwlocks > and the lock was taken on read when traversing (around 2.4.0-pre1 > or so). Then, it turned out that ANK's networking wants to delete > notifiers from a running notifier. So, someone else (Linus?) removed > the taking the lock for read. Now we have an rwlock which is only > taken for write :-) Maybe it makes sense to look at 2.3.99. > It's all so shrouded in the mysts of time, that maybe it's easier > to lock outside now.
I got your message, and rewrote the notifier code (basically took a local copy of it, and added proper locking). Look on the list for my followup patch. thanks, greg k-h ------------------------------------------------------- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl _______________________________________________ [email protected] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
