> > Hmm, so you actually think it's a feature that driver refcounting > > bugs can cause random flakiness for a couple years before > > they're finally tracked down? > > This is the wrong way of doing that. Poisoning is the best of catching > reference counting problems.
Not "best" ... as you commented, any "catch" would be after the real bug escaped surveillance. And poisoning is only found on systems built for debugging (and not even all of them) ... plus even then, there are failure modes it can't isolate. (Start with memory getting reused sooner in some cases than others ... :) > It's hardly simple. To be able to use the reference counting that you've > changed to, my uhci_free_dev has to be very complex to be able to catch > all of the corner cases. If you clear qh->dev after the associated urbs are unlinked/completed (the issue discussed so far), that latent bug would vanish. Unless there's some additional issue lurking, I'd suspect you could just keep that NOP version of uhci_free_dev(). - Dave _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
