[Ftdi-usb-sio-devel] Continuous stream of characters sent by ftdi_sio driver.

[G.W. H. <ft...@ju...>]

Hi there,

Recently I've been having trouble with a number of Seagate 'Barracuda'
hard drives, and one of the issues appears to be resolvable by means
of sending commands to the drive through a TTL-level serial interface
which apart from the voltage levels otherwise somewhat adheres to the
EIA/TIA-232 standard.  This USB/TTL-level serial interface converter
is available from a firm called KMtronic:

http://sigma-shop.com/product/112/ft232bm-bl-seagate-barracuda-7200-11-firmware-fix-tool-complete-usb-powered.html

It seems to work for most people who use it, but most people use what
I shall call a different operating system. :)

Under Linux, lsusb -v gives:

Bus 004 Device 005: ID 0403:6001 Future Technology Devices International, Ltd FT232 USB-Serial (UART) IC

The current Linux kernel driver for this device is ftdi_sio.  When I
installed it in a Slackware box with a 3.8.3 kernel it failed with the
symptoms below.  Normally I use minicom for serial comms but the same
problem appears with another serial terminal package.  The same also
when I use the 'generic' driver instead of ftdi_sio.

1. The driver appears to be sending a constant stream of characters to
the computer (i.e. to the machine with the USB port).  The characters
are alternatiely 0x01 and 0x60.

2. Activity is seen on the LEDs on the USB-serial device itself only
when expected, i.e. when typing characters at the terminal (Tx LED)
and receiving them from the connected equipment (Rx LED).

3. When the converter is connected in a loopback arrangement, all the
characters sent via the serial terminal application (minicom) to the
device are echoed back properly from the serial output to the input as
expected, but they are interspersed with the continuous stream of 0x01
and 0x60 characters.

I'm pretty sure that the hardware is working OK.

The version of the driver available on the FTDI website is very old,
and contains a comment which refers to symptoms such as I am seeing.
Unfortunately this version of the driver does not compile for 3.8.3
and before I delve more deeply into the code I'm hoping that someone
with more intimate knowledge of the driver will be able to point me in
the right direction to fix it.

--

73,
Ged.