> Frank Gockel <gockel@sent13.uni-duisburg.de> wrote:
> [...]
> This is the third release of my 16550C autoflow diff for 2.0.xx kernels.
> It was created against 2.0.35 and may not work for other kernels. It is
> expected to fail under latest 2.1.xx kernels (not tested).
Ah, another UART, life was so boring without it (grin).
Is it at least compatible with the SSI equivalent product,
something like SSI7550 ? just checked, no, even if TI buyed SSI.
By the way, if you like to write some code, you should switch
to 2.1, more exactly 2.3 because of the 2.1 code freeze, the
serial.c file has evolved since 2.0.xx .
In 2.1.xxx, 16650 and 16750 are recognised, their deep FIFO and
hardware autoflow are activated.
Here is my humble comments:
- I do not see where you manage the special meanning of RTS
in Autoflow mode.
(from http://www-s.ti.com/sc/psheets/slls177e/slls177e.pdf
page 31 table 8)
- Your detection of 16550C conflict with 16650 and 16750 in 2.1,
those can modify this "autoflow control bit", but this bit
has not necessarily the same meanning...
- The use of autoflow will be better controlled by some
run-time parameter instead of "#define"
- If you are using a NULL modem with one of your line,
the DCD line is connected to CTS and CTS should not
be modified for the complete session.
Bye,
Etienne.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/