psarc/2009/197 - UFTDI - fast-track

[James Carlson]

Tim Marsland writes:
> This case asserts that, moving forward, the default behaviour
> for "off-board" serial ports should be the same as "on-board"
> serial ports -- the distinction being particularly ambiguous
> in the case of USB simulations of RS232 devices.

I can understand why this is the default for possibly broken hardware
(devices ordinarily lacking standard modem control lines), and when we
discussed the issue, I had thought that was the context for the
change.

However, I don't think it makes sense as a general rule for all
off-board serial ports.  The only reason it's the default for on-board
ports is because of the common (and legacy) usage of /dev/term/a as a
console port with a three-wire connection to an ASCII terminal.  In
fact, given the legacy nature, I'd be much more in favor of saying
that going forward all ports -- both on-board and off-board -- should
*NOT* have the "ignore-cd" property set by default.

I'd settle for saying that ignore-cd=1 is the default for UFTDI,
because of an unusual risk of broken hardware in this particular usage
case (as I thought we'd discussed), but I don't think it's right as a
general rule.  It provides for strange operation at best, and security
problems (such as disconnected lines being left logged in) at worst.

(The better answer is for the driver to supply the default: if the
hardware is known to be deficient or otherwise "special," then using
ignore-cd=1 as the default is fine.  If it's known to be a properly
functioning serial port with a reasonble complement of modem control
lines, then it should be ignore-cd=0.)

-- 
James Carlson, Solaris Networking              <james.d.carlson at sun.com>
Sun Microsystems / 35 Network Drive        71.232W   Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757   42.496N   Fax +1 781 442 1677