On Mon, Aug 21, 2000 at 04:30:04PM -0700, Greg KH wrote:
> On Mon, Aug 21, 2000 at 04:13:00PM -0700, Dunlap, Randy wrote:
> >
> > usblp_read() and usblp_write() already use separate urbs
> > for read vs. write and for each instance of the lp (minor).
> > Can you give me a brief explanation of the problem?
> > (and I'll look at David's dc2xx patch).
>
> >From what I understand, copy_to_user can sleep, then (for the usblp_read
> function) the read can be called again, the data can be written to the
> urb and submitted, then the previous copy_to_user wakes up, and tries to
> resubmit the urb again with different data this time. Not good.
>
> Same thing for the usblp_write function.
IMHO that can't happen, because these devices are single-open only, that
is only one process can be using then at once. So I can't imagine how it
would be able to sleep (in copy_to_user) and reenter the read/write
routines both at once.
-- Vojtech Pavlik SuSE Labs - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Aug 23 2000 - 21:00:06 EST