Re: Strange problem with tty layer

[Paul Fulghum <paulkf@microgate.com>]

Lennart Sorensen wrote:
> I have confirmed that the driver does in fact receive all the
> characters, and that they are correct, and that they are being passed to
> the tty layer using tty_insert_flip_string, and that it returns that all
> the characters have been passed to the tty layer.  The user space
> application however still doesn't see the last few characters (when it
> fails).
> 
> The problem seems to occour every few hours of testing on a 266MHz Geode
> SC1200.  When I change the clock to 133MHz, it happens every few minutes
> instead (so much more frequently).  I suspect there is some race
> condition that allows the tty layer to not get around to processing all
> the data in the buffer, even when asked for data by the application
> (which is waiting on the serial port using select, with a 4s timeout).

In 2.6.16 the tty buffering pushes data to the line
discipline without regard to tty->receive_room.
If the line discipline can't keep up, the data gets dropped.
I observed this data loss at higher speeds when
placing the system under heavy load.

2.6.18 added code to respect tty->receive_room.

This may or may not be your problem, but you should
be able to check by adding a conditional printk
to drivers/char/tty_io.c:flush_to_ldisc()

If tty->receive_room is less than the size of the buffer
passed to disc->receive_buf() then you are losing data.

-- 
Paul Fulghum
Microgate Systems, Ltd.