On Mon, 06 Dec 2004 01:09:22 -0500, Ryan Reading <[EMAIL PROTECTED]> wrote:

> My first question was why they n_tty.c:write_chan() function does *NOT*
> call schedule() for tty_driver:write() calls that write all of the data
> requested.  From a userland perspective, this means that a write()
> cannot be guaranteed to happen successfully because we may still be
> waiting on an interrupt to complete the usb_submit_urb().

It's no different from any UART with a transmit FIFO.

> My second question is, if the implementation of n_tty.s:write_chan() is
> correct, why don't usb drivers that implement the tty_driver interface
> block write() calls until the completion of the usb_submit_urb()??  It
> seems to me that userland protocols that rely on serialized I/O will
> have a real problem with this implementation.

I don't see how serialized I/O figures here, as long as writes appear on
the wire in order. What is your definition of "serialized I/O" is?

-- Pete


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now. 
http://productguide.itmanagersjournal.com/
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel

Reply via email to