DL(4) NetBSD Programmer's Manual (VAX Architecture) DL(4) NNAAMMEE ddll - DL11/DLV11 serial device driver SSYYNNOOPPSSIISS ddll00 aatt uubbaa?? ccssrr 00117766550000 ddll11 aatt uubbaa?? ccssrr 00117766551100 ddll22 aatt uubbaa?? ccssrr 00117766552200 ddll33 aatt uubbaa?? ccssrr 00117766553300 DDEESSCCRRIIPPTTIIOONN The ddll driver controls a DL11-compatible asynchronous serial card, and probably things compatible with it. A four-port DLV11-J will appear four times in the device list, as the ports look like separate cards to the driver. The ddll driver provides the normal interface described in tty(4), but many of the configuration calls are unsupported, since their functions are handled by jumpers or switches on the serial card itself. Calls re- lated to modem-control lines are also ignored, since these cards lack them. There's a chance this driver might also work with an LP11, an LPV11 or even a PC11, but it hasn't been tested. FFIILLEESS /dev/ttyJ? Special files for communicating with dl devices. DDIIAAGGNNOOSSTTIICCSS ddll%%dd:: rrxx oovveerrrruunn.. The character in the recieve buffer wasn't read before the next character arrived, and has been lost. ddll%%dd:: ssttrraayy rrxx iinntteerrrruupptt.. The driver was called to service a receive in- terrupt, but there was nothing for it to read. SSEEEE AALLSSOO tty(4) HHIISSTTOORRYY The ddll driver was written for NetBSD 1.2B. AAUUTTHHOORR Ben Harris BBUUGGSS The DL11 and friends only have single-character receive and transmit buffers, so an interrupt is generated for every character received or transmitted. Attempting to receive data at even moderately high rates will cause rx overruns. Fast transmission seems to be fine though. There is no support in the driver for the paper-tape reader on an LT33 attached via a DLV11-KA or similar. The overrun message is logged in the interrupt routine itself, which will probably just make the problem worse. The CSR printed on startup is that of the receiver, while the interrupt vector is that of the transmitter. In order to determine the card's interrupt vector, the driver sends a NUL to each port. This may confuse things attached to them. The driver has so far only been tested on a DLV11-J. It may or may not work on the other cards it claims to support.