From: Georg Potthast 2 <nospam@georgpotthast.de>
To: Paul Osmialowski <pawelo@king.net.pl>
Cc: Derek Johansen <djohanse678@gmail.com>,
"Marc-F. Lucca-Daniau" <mfld.fr@gmail.com>,
ELKS <linux-8086@vger.kernel.org>
Subject: Re: Obsolete documentation?
Date: Mon, 17 Feb 2020 15:47:23 +0100 (CET) [thread overview]
Message-ID: <718029432.180460.1581950843768@com4.strato.de> (raw)
In-Reply-To: <alpine.LNX.2.21.1.2002170001320.7728@localhost.localdomain>
Hi Paul,
I recall that DOS used the BIOS to send and receive data via the serial port. This resulted in a usable speed of 1200 baud max. Lower speeds were more reliable. Therefore all communication programs used interrupt driven data input and no DOS or BIOS calls. I assume ELKS also uses the BIOS and therefore you are limited to 1200 baud in case of a 8086 processor and have to expect missing data then.
Georg
> Paul Osmialowski <pawelo@king.net.pl> hat am 17. Februar 2020 um 00:26 geschrieben:
>
>
> Hi Georg,
>
> I'm using null-modem cable I've made myself and been using it for years
> (it corsses-over Rx/Tx and hardware control lines). It works nicely with
> Telix running under FreeDOS on my XT and minicom on 'big' Linux, at speed
> 115200, which is the top speed for 8250 chip on the XT side.
>
> With ELKS it's not that nice. I've managed to configure getty in
> /etc/inittab to use /dev/ttyS0 at the default speed (9600bps), and it
> kinda worked, except it was losing bits from time to time. I'd removed
> getty line from /etc/inittab for other experiments, then I've tried to
> 'cat' some text files between ELKS and Linux (and vice versa), and
> although a text sent from ELKS to Linux looked good, the other way round
> it was disaster. I've tried different speeds (using 'stty' on both sides).
> Finally, it looked almost OK at 1200bps, but still I had an impression,
> bit-banging serial port on my old ZX Spectrum +3 offered better
> communication stability at 4800bps than this!
>
> Eventually, I've managed to establish SLIP connection between ELKS and
> Linux, with telnetd running on ELKS, and I could open telnet connection
> that worked for a while... not too long though. I guess, serial connection
> support must be looked upon in ELKS before doing experiments with SLIP.
>
> Trying to follow your instructions, I've encountered some problems with
> the 'ifconfig' line:
>
> - there's no 'up' param in your example; usually I'm adding it after the
> interface name (e.g. 'ifconfig sl0 up ....'),
>
> - the destination address option is different in various versions of
> ifconfig; in your example it was 'pointtopoint', but my 'ifconfig' didn't
> like it; turned out, in my case it should be 'dstaddr' (which some googled
> pages listed as obsolete).
>
> Thanks,
> Paul
>
> On Sun, 16 Feb 2020, Georg Potthast wrote:
>
> > ELKS does NOT have a special version of SLIP. Therefore you can communicate
> > with any Linux host provided you manage to configure this. I would recommend
> > to follow my instructions first to get an understanding how to set this up.
> > Then make a serial cross over connection cable and test that using two
> > terminal programs on each side. Getting the wires linked correctly and
> > providing the hardware signals is not trivial for someone who has not done
> > that before. If you can send across what you type you can start configuring
> > SLIP over this line.
> >
> > Georg
> >
> > -----Ursprüngliche Nachricht----- From: Paul Osmialowski
> > Sent: Sunday, February 16, 2020 7:37 PM
> > To: Georg Potthast
> > Cc: Paul Osmialowski ; Derek Johansen ; Marc-F. Lucca-Daniau ; ELKS
> > Subject: Re: Obsolete documentation?
> >
> > Hi Georg,
> >
> > Will it connect to SLIP/CSLIP interface as implemented in normal Linux
> > kernel (CONFIG_SLIP)?
> >
> > Thanksm
> > Paul
> >
> > On Sun, 16 Feb 2020, Georg Potthast wrote:
> >
> > > I wrote a SLIP documentation which shows how to setup a SLIP connection
> > > between an ELKS system running in QEMU und the host where QEMU is running.
> > > You
> > > could send data from the host to the ELKS system and vice versa. This is
> > > this
> > > document:
> > > elks/Documentation/html/user/setup_slip.html
> > > To me this seemed easier to set up instead of two ELKS systems connected
> > > with
> > > a cross-over serial cable.
> > >
> > > Georg
> > >
> > > -----Ursprüngliche Nachricht----- From: Marc-F. Lucca-Daniau
> > > Sent: Sunday, February 16, 2020 6:05 PM
> > > To: Paul Osmialowski ; Derek Johansen
> > > Cc: ELKS
> > > Subject: Re: Obsolete documentation?
> > >
> > > Hello Derek and Paul,
> > >
> > > ELKS is indeed still SLIP capable :
> > > https://github.com/elks-org/elks/issues/304
> > >
> > > One has just to change the command line of 'ktcp' to remove Ethernet and
> > > put back SLIP.
> > >
> > > Documentation is quite outdated. Georges made an effort to document
> > > Ethernet and some other things while debugging networking features, but
> > > there are still many legacy documents that need to be reorderer and updated.
> > >
> > > Thanks,
> > >
> > > MFLD
> > >
> > >
> > > Le 16/02/2020 ? 12:08, Paul Osmialowski a écrit :
> > > > So what happened to SLIP support in ELKS? 8-bit ISA ethernet cards
> > > > (with RJ-45 connector) are very rare (I have only one of them, work >
> > > nicely
> > > > under FreeDOS and in theory it should be supported by ELKS's ne2k >
> > > driver,
> > > > but it isn't). Within my one more XT machine to play with ELKS, serial
> > > > port is the only means of external communication. PPP/IP or SLIP is
> > > > something ELKS should definitely have IMHO.
> > > >
> > > > Cheers,
> > > > Paul
> > > >
> > > > On Sat, 15 Feb 2020, Derek Johansen wrote:
> > > >
> > > > > Does elks/Documentation/text/networking_guide.txt make
> > > > > elks/Documentation/text/networking.txt obsolete? The latter in
> > > > > Section 2 says ELKS only supports SLIP connections. I don't think
> > > > > this is still true?
> > > > >
> > >
> >
next prev parent reply other threads:[~2020-02-17 14:47 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-16 4:51 Obsolete documentation? Derek Johansen
2020-02-16 11:08 ` Paul Osmialowski
2020-02-16 17:05 ` Marc-F. Lucca-Daniau
2020-02-16 18:10 ` Georg Potthast
2020-02-16 18:37 ` Paul Osmialowski
2020-02-16 21:17 ` Georg Potthast
2020-02-16 23:26 ` Paul Osmialowski
[not found] ` <CALgV52jgt1U0MNAxEy6jsDykdkMHGGopBxT4o6itW4UzWrXx5A@mail.gmail.com>
2020-02-17 13:10 ` Paul Osmialowski
2020-02-17 14:47 ` Georg Potthast 2 [this message]
2020-02-17 15:07 ` Paul Osmialowski
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=718029432.180460.1581950843768@com4.strato.de \
--to=nospam@georgpotthast.de \
--cc=djohanse678@gmail.com \
--cc=linux-8086@vger.kernel.org \
--cc=mfld.fr@gmail.com \
--cc=pawelo@king.net.pl \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox