printer half working. Need help!

[Vernon Schryver]

All of the minor devices for a given major are handled by the same driver.
The ttyd*, ttym*, and ttyf* minors for the 6/8-port (CDSIO) and integral
ports differ:
   ttyd*    use pins 2,3, &7.  RTS and DTR are high while the port is open

   ttym*    use 2,3,7,8, and 9/20  (9 on the 9-pin connector, 20 on a DB-25). 
		DCD and DTR are used in the standard RS-232 way.  An open(2)
		will not be completed unless NDELAY is set or DCD (pin 8)
		is tree.  DTR is high or true from start of the open until
		the port is closed.
				

   ttyf*    use 2,3,4,5,7,8, and 9/20.   Beside the standard DCD/DTR
		dance, output is stopped while CTS is false.  In addition,
		the de facto standard "hardware flow control" use of 
		RTS is done, with RTS low meaning the computer (often
		the application) is not ready for more input.

All of this is documented in chapter 10 of the new (IRIX 3.3 & I think 3.2)
"IRIX System Administrator's Guide", Document Version 3.0, Document
Number 007-0603-030.

Missing output to a printer sounds like missing flow control.  One familiar
cause is that the device has not been configured to generate flow control
signals, either XON/XOFF or a wire.  Ancient devices from the 1970's and
1980's often used DSR or DTR for "hardware flow control." I think I
remember some (e.g. TI-800's) that optionally used odd pins like 18 for
flow control.  Creative cabling can be useful in such cases.  You can hook
pins 2,3, &7 between the device and computer, loop the computer's 8 to its
9, connect the device's flow control signal to the computer's CTS input,
and use /dev/ttyf*.

A "break out box" is a vital tool for such games.  A "Y-cable" and a dumb
terminal can make an effective and cheap "line monitor.




Vernon Schryver,    vjs at sgi.com