lau@kings.wharton.upenn.edu (Yan K. Lau) (11/28/89)
Has anyone been able to get lpr printing to work with the parallel port on the Apollo SPE board? I've setup /etc/printcap with /dev/pio. Lpr queues jobs but nothing gets printed. The jobs just sit in the queue. Lpd is running. I'm trying to hook up a HP Laserjet if this makes a difference. Any ideas? Yan. )~ Yan K. Lau lau@kings.wharton.upenn.edu The Wharton School ~/~ 128.91.11.233 University of Pennsylvania /\ Darker grows the moon And shadows steal across the prison of my room
jaeger@madnix.UUCP (Jay Jaeger) (11/30/89)
In article <17339@netnews.upenn.edu> lau@kings.wharton.upenn.edu (Yan K. Lau) writes: > >Has anyone been able to get lpr printing to work with the >parallel port on the Apollo SPE board? I've setup /etc/printcap >with /dev/pio. Lpr queues jobs but nothing gets printed. The >jobs just sit in the queue. Lpd is running. I'm trying to hook >up a HP Laserjet if this makes a difference. Any ideas? > > We at WI DOT also had tons of trouble trying to make parallel printers work under SR9.7. We finally gave up. I think there are some very basic problems involving parallel ports, interrupts and Apollo device drivers (in the kernel, not the ones that are really just filters that Apollo sometimes calls "drivers"). Basically, I think that /dev/pio is being asked to do an impossible task - PC parallel ports do not do well being interrupt driven - one look at all the trouble the MINIX folks have had in that area should convincie you. Very best of luck. (Sorry I don't have a signature line yet, I am: Jay R. Jaeger State of Wisconsin, Dept. of Transportation email (for now): madnix!jaeger (I HOPE) -----------------------------------------------------------------------------
dbfunk@ICAEN.UIOWA.EDU (David B Funk) (12/01/89)
In article <17339@netnews.upenn.edu> lau@kings.wharton.upenn.edu (Yan K. Lau) writes: > >Has anyone been able to get lpr printing to work with the >parallel port on the Apollo SPE board? I've setup /etc/printcap >with /dev/pio. Lpr queues jobs but nothing gets printed. The >jobs just sit in the queue. Lpd is running. I'm trying to hook >up a HP Laserjet if this makes a difference. Any ideas? > You should be able to use Lpd to drive your printer if you can hook it to the serial port. I'm not sure that it'll be worth the bother to connect it to the parallel port. Yes SPE does provide a streams interface to the parallel port, but you may also need to use a special "spe_$pio_set_mode" call to configure the handshake options. This is OK within a new program that you're writing, but an existing program like Lpd doesn't know about this suff. It may not even be worth it from a speed stand point. I took one of my prsvr HP LaserJet drivers and converted it from sio to pio output as a test. I found that the SPE parallel I/O was about 20% faster than the serial line at 9600 baud, and about 80% slower than the serial line at 19200 baud. there was also lots more cpu overhead on the node when driving the pio port. This was under sr9.7 with SPE v1.3, I've not tried it under sr10.x. Dave Funk
krowitz%richter@UMIX.CC.UMICH.EDU (David Krowitz) (12/01/89)
I've written GPIO drivers for the IBM-PC/XT parallel card (which is hardware compatible with the SPE parallel port) and the Ikon 10092/10097. I run the XT card using programmed I/O and get roughly 60,000 characters per second through it to our Shinko color printers (running on a DN3000, on a DN3500 the rate goes up to roughly 100,000 bytes/sec). It seems like the SPE driver must be trying to use interrupt driven I/O -- which is a *very* bad idea with a character-at-a-time device. I will be looking into doing a streams manager interface for these GPIO drivers so that the parallel port can be simply referred to a /dev/lpt in the print-server config file ala Michael Lampi's suggestion. -- David Krowitz krowitz@richter.mit.edu (18.83.0.109) krowitz%richter.mit.edu@eddie.mit.edu krowitz%richter.mit.edu@mitvma.bitnet (in order of decreasing preference)