> First, instead of relying on port->port.count we should have a private > port->open_count field. That way we can reliably be certain to call > the subdriver's open & close methods in strict succession. It is called ASYNC_INITIALIZED and is used by the core and most of the other drivers - except USB. I've just been fixing the USB serial layer to use it. The basic idea is that ASYNC_INITIALIZED is set when the hardware is initialized and cleared when it isn't. It's a bit flag manipulated by set/clear/test bit so can be safely snooped by other code. I've done that one for the USB stack because Oliver Neukum needs to be able to check that for the resume paths on option so I'll just push it up the queue a bit. > Second, if tty_port_block_til_ready fails then we have to call the > subdriver's close method. Ok. I'll drop in the driver close call in the next patches I send out. Alan -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
