*BSD News Article 21246


Return to BSD News archive

Xref: sserve comp.os.386bsd.questions:5245 comp.os.386bsd.development:1233
Path: sserve!newshost.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!elroy.jpl.nasa.gov!swrinde!cs.utexas.edu!uunet!math.fu-berlin.de!irz401!uriah!not-for-mail
From: j@bonnie.tcd-dresden.de (J Wunsch)
Newsgroups: comp.os.386bsd.questions,comp.os.386bsd.development
Subject: Re: Epson LQ printer - Can it be done?
Date: 21 Sep 1993 13:27:48 +0200
Organization: Textil Computer Design GmbH Dresden, Germany
Lines: 79
Message-ID: <27mofkINNjo5@bonnie.tcd-dresden.de>
References: <TONY.93Sep9182513@marge.apana.org.au> <1993Sep11.124630.22284@gmd.de> <279b1mINNjgb@bonnie.tcd-dresden.de> <1993Sep16.115459.24351@gmd.de>
NNTP-Posting-Host: bonnie.tcd-dresden.de

veit@mururoa.gmd.de (Holger Veit) writes:

>j@bonnie.tcd-dresden.de (J Wunsch) writes:
>|> 
>|> Please accept that this IS A BUG in *BSD, not in the cables. The bug has

>If so, please explain me the following things:

>2. If this is a bug, and everyone agrees that it is, why the hell hasn't
>   someone digged into that and tried to correct this? According to 

Since it's a very nasty thing. Appears only on some machines. Only
sometimes. - But not only at mine. I've already seen some other postings
saying `the lpa driver doesn't recognize', `i have to detach the printer
to be detected' and so on. You've noticed it, and you've answered some
of them, too - but only with your `check the cable' message.

Probably, lotta people gave up here. Went to linux, to serial printers
or what the hell... Other people may have succeeded with another
printer card. -- But i don't think it's good to leave it its way, since
my h/w configuration didn't change but worked before 0.2.4.

>   ... But obviously it is
>   more fun to change db formats, a.out formats to make everything
>   different than to fix essential things.
agreed.

>3. If this worked with pre-0.2.3 lp drivers, is it really so difficult to
>   compare the working version with the apparently buggy "innovation", and
>   analyze the effect? Apparently some changed code caused this, and if 

Since - as i said - i'm still using the _same_ lp driver, which also
sucks now, it's not so obvious where to search. At least, the lp* drivers
seem to be *not* the real problem.

>   it is not the lp driver, possibly another side effect has been built in.
...
>   I think the (mostly correct) improvement of interrupt handling by bde 
>   caused this (and thus triggered misbehavior of some hardware), but then
>   the old code shouldn't work any longer...

Maybe, but i didn't really track it down. (My time's limited, sigh.)
Will look here next time.

>4. I don't know about your printer, but sometimes it appears to be a
>   flaky or non standard hardware interface of the printer than
>   really a software bug in *BSD (Don't tell me it works fine with DOS).

But i tell you it works fine with DOS. Even works fine with an interrupt-
driven driver there (i used to have due to some race condition with
some grafx software).
The printer is an Epson LQ550, not the best, a bit auld, but also
no $$10 hardware. Worked fine from the old 386bsd/0.0 days on.
Btw., with the same cable all the time.

>   So, did you exchange cables, printer, printer port, tried on *different*
>   systems?
(Currently) only a single system fails. (But i don't have dozens thereof)

>|> printer port. Yes, indeed, some{thing,one} is writing some byte to the
>|> printer port - it ain't the lp driver, though!

>So then, rename the /dev/lp to something like /dev/myprinter
>(so normal apps should no longer find a printer where they can send 
>bytes to),
NONONO. It's not a program that writes it. A normal program has no
opportunity (via the /dev/lp interface) to corrupt some of the status
lines. But: the SELECT line gets inactive after some time.

>|> I have no idea where to look for this...
>Should be enough stuff to check in this letter.

Not much more. Besides the suspicion of the intr code...

-- 
in real life: J"org Wunsch |   )  o o  | primary: joerg_wunsch@tcd-dresden.de
above 1.8 MHz:   DL 8 DTL  |    )  |   | private: joerg_wunsch@uriah.sax.de
                           | . * ) ==  |
          ``An elephant is a mouse with an operating system.''