On Thu, May 16, 2002, David Brownell <[EMAIL PROTECTED]> wrote:
> > I think the patch that David proposed to move deallocate() into
> > usb_disconnect() should go into 2.5 and 2.4 since it's needed to make the
> > code safe.
>
> ... in the face of buggy drivers (which we know happen with
> unfortunate inevitability :). Yes.
That's not why. It's because deallocate() right now can be called from
any context, buggy or not, which is not what sohci_free_dev wants.
> In fact this patch would likely have avoided all the flamewars,
> since it provides the essential guarantee by decoupling the
> memory management ("refcounting") from the cleanup that
> relates to the hardware going away.
It should be renamed then.
However, it still doesn't solve how complex sohci_free_dev is.
> > That will be immediately make the code safe at the cost of some
> > complexity in some HCD's.
>
> Nope, no new complexity.
No new complexity, but still some unneeded complexity.
JE
_______________________________________________________________
Have big pipes? SourceForge.net is looking for download mirrors. We supply
the hardware. You get the recognition. Email Us: [EMAIL PROTECTED]
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
