PLIP - Parallel IP - does it still work ?

Julian H. Stacey jhs at berklix.org
Tue Dec 11 12:30:40 PST 2007


Oliver Fromme <olli at lurza.secnetix.de> wrote:

> FWIW, I also noticed that PLIP seems to work best when
> both machines are about the same speed.  If you can't
> get it to work between a very fast machine and a slow
> one, then I suggest you try a different machine (if
> possible) that is of similar speed as the other one.

Thanks. I started with a similar slow server "lapl" which the 7.0BETA3
installer on install client "norb" failed to contact, then I switched
to a fast server "fire" that also failed.

Hosts "lapa" & "lapl" below work, so a reference to test PL-IP links with:
 HOST  RELEASE   PROCESSOR
 fire  6.2       AMD Athlon(tm) 64 Processor 3000+ (2010.06-MHz K8-class CPU)
 lapa  4.11      Pentium II/Pentium II Xeon/Celeron (267.27-MHz 686-class CPU)
 lapd  7.0-BETA3 Pentium/P55C (166-MHz 586 CPU) (no ppc yet, config prob)
 lapl  4.11      Pentium/P55C (120.27-MHz 586-class CPU)
 loft  6.2	 Pentium II/Pentium II Xeon/Celeron (334.09-MHz 686-class CPU)
 norb  7.0-BETA3 Intel Pentium (P54C) (586-class), 133.65 MHz, id 0x52c

Links: lapa-loft, & fire-lapl dont work.
  ( Host fire (X server) display same sticky mouse latency I've
  seen previously on a working FreeBSD plip/slip though, so some
  activity on that cable, it's trying even if not a working IP link.
  (No lpd of course, 3 PLIP cables, all tested OK on lapa-lapl)

I think PLIP in much or all of FreeBSD-[5-7].* is defective.
	http://www.freebsd.org/cgi/query-pr.cgi?pr=113856
	http://berklix.com/~jhs/src/bsd/fixes/FreeBSD/src/gen/share/man/man4/lp.4.REL=ALL.diff
	http://berklix.com/~jhs/src/bsd/fixes/FreeBSD/src/gen/share/man/man4/plip.4.send-pr.ignore
	http://berklix.com/~jhs/src/bsd/fixes/FreeBSD/src/jhs/sys/dev/ppbus/if_plip.c.REL=ALL.diff
Probably time `man plip` is marked as Dead, until fixed.


"Poul-Henning Kamp" <phk at phk.freebsd.dk> wrote:

> PLIP is sensitive to cpu clock speeds, you may have to extend the
> loops somewhat in order to give all the EMC/EMI protection time to
> pass the signals.

Thanks, Some time I'll hope to look in /usr/src/sys/ dev/ppc & modules/ppc
if no one else does.

-- 
Julian Stacey. Munich Computer Consultant, BSD Unix C Linux. http://berklix.com
	Ihr Rauch = mein allergischer Kopfschmerz. Dump cigs 4 snuff.


More information about the freebsd-current mailing list