info-kermit@ucbvax.ARPA (07/13/85)
From: Frank da Cruz <SY.FDC@CU20B.ARPA>
Info-Kermit Digest Fri, 12 Jul 1985 Volume 3 : Number 3
Departments:
ANNOUNCEMENTS -
Another Release of C-Kermit 4C for Unix, VMS, and Macintosh
Commodore-64 Bootstrap Program in Fortran
Assembling Apple II Kermit (AP2KER) with Apple Assembler?
New Apple II Cross Assemblers
OLD QUERIES ANSWERED -
Re: Kermit on MicroVAX-I
Re: Z100 Communications Program Query
NEW QUERIES -
IBM 3270/PC and Kermit?
----------------------------------------------------------------------
Date: Fri, 12 Jul 85 16:58:22 EDT
From: SY.FDC@CU20B (Frank da Cruz)
Subject: Another Release of C-Kermit 4C for Unix, VMS, and Macintosh
Yet another release of C-Kermit 4C, this one is 4C(056), for Unix, VAX/VMS,
and the Apple Macintosh.
Major differences from 4C(055), affecting all systems:
. Various file transfer performance improvements.
. Another bug with 8th-bit-prefixing negotiation fixed.
. Some bugs fixed concerning interrupted file transfers.
. Incoming Z (EOF) packet now ack'd only after file successfully closed.
. Allowance made for ACK to F packet containing alternate file name.
. "Blank lines" in packet stream no longer NAK'd.
. Test for echoed packets fixed.
. Input buffer now flushed only after desired packet is read.
. Many minor fixes and cleanups.
Unix and VMS only:
. "statistics" and transaction log now include timing information.
. "set incomplete {keep,discard}" is now implemented.
. "show parameters" redesigned and has some inaccuracies fixed.
. "echo" now accepts \ooo octal escapes, e.g. "echo foo\007bar" will now beep.
. "set prompt" now accepts argument in doublequotes to allow blanks.
. Command parser now accepts comment lines starting with "%".
. It is now possible to exit protocol at any time by typing ^A^C^C<CR>.
. Manual chapter updated to reflect above changes.
Unix only:
. 4.xBSD performance improved by doing nonblocking reads, own buffering
(Herm Fischer's Sys III/V code adapted to work for BSD).
VMS only:
. Version numbers should now be stripped from outbound file names.
. Improved build procedure (CKV*.COM).
Macintosh:
. No Mac-specific changes were made, but the changes made to the system-
independent protocol modules should fix a couple problems that reportedly
prevented all C-Kermits from communicating with systems that send "blank
lines" between packets, with systems that echo packets, and with Kermits
that know about 8th-bit prefixing but refuse to do it. The new Mac Kermit
release is numbered "0.8(33)".
The new release has been tested under 4.2bsd, Pro/Venix V1, and on the
Macintosh against normal systems like DEC-20's and VAXes as well as against
IBM mainframes both in line mode and with full screen 3270 protocol
conversion (thru Series/1). The VMS version has not been tested (feedback
welcome).
Thanks to Larry Afrin, Gary Algier, Todd Booth, Kelvin Nilsen, Ken Poulton,
Dan Schullman, Ed Schwalenberg, and others for reporting problems and/or
suggesting fixes since the last release of C-Kermit 10 days ago.
The new files are in K2:CK*.* on CU20B, available via anonymous FTP.
The updates are listed in greater detail in the files K2:CK*.UPD. The files
K2:CK*.BWR list known bugs and restrictions. K2:CKUKER.DOC is the new
Kermit User Guide chapter for Unix Kermit (not yet integrated into the
monolithic Kermit User Guide, KER:KUSER.DOC).
Since C-Kermit continues to change, it is not recommended that you get these
files unless:
(a) You are having problems with your present version that might be fixed
by the changes listed above;
(b) You are doing development work based on the C-Kermit source (always try
to work from the latest sources!).
It is expected that these new files, along with others recently announced,
will be available via uucp at okstate shortly after okstate comes back up on
July 22 or thereabouts, and will also be available on BITnet via KERMSRV at
CUVMA probably next week.
As usual, send comments, suggestions to Info-Kermit@CU20B.
------------------------------
Date: 10:23:48 07/11/85 (85.192)
From: Jeff Balvanz <GM.JLB @ ISUMVS>
Subject: Commodore-64 Bootstrap Program in Fortran
This is a Fortran version of the download program C64BOOT designed to
bootstrap the Commodore-64 version of Kermit from our VAX system. It should
work with minor modifications with any system running Fortran-77.
[Ed. - Thanks, now we have one in each of Fortran, C, Simula, and Clu --
quite a collection -- in KER:C64BOOT.* (.FOR, .C, .SIM, .CLU; pick your
favorite.)]
------------------------------
Date: Fri, 12 Jul 85 08:52:18 MST
From: slesh@FTH-1
Subject: Assembling Apple II Kermit (AP2KER) with Apple Assembler?
Assembling and using the Apple Assembler/Editor version of Kermit
is NOT a straightforward proposition. (The following comments are based
upon as little expertise with Apple 3.3 DOS and the assembler as I could
acquire in the process of getting an Apple DOS Kermit.) A few of the little
quirks I have discovered THE HARD WAY follow:
1- if you obtain the source ( ap2ker.asm ) using a
communications program which runs under another operating system
and permits text files larger than 32K (e.g. CP/M 80), you will not
be able to convert the resulting file to an Apple text file without
splitting it. My Applicard file conversion utility ADOSXFER
converted the first 32K all right then went right on cheerfully
whirring disks and flashing text on the screen. When I attempted to
assemble the resulting product there were 81 errors.
2- the Apple Assembler documentation mentions something about
'chaining' - "CHN" - but I couldn't get that to work. What did
work was naming two source files on the "asm" command line. I
couldn't find anything in the documentation to indicate WHY it
worked. (The binary file takes the name of the second source file
named - an undocumented feature of the Apple Assembler?)
3- the 'ap2ker' version does NOT have some of the features
documented in the DEC 20 version (e.g. talking to servers, setting
slots or defining keyboard type).
A little help from anybody out there who really knows what's going
on would be appreciated.
[Ed. - AP2KER is a "hand crafted" translation of the original Stevens
Institute of Technology "CROSS" cross-assembler source into Apple assembler,
done by Olaf Pors of the University of Virginia. It is, indeed, based upon
an older version of Apple Kermit; the CROSS version continued to change
after Olaf contributed this version, and Olaf made a few changes during
translation that may or may not have shown up in the "master copy". CROSS,
for those who don't know, is a general-purpose cross assembler written in
PDP-10 assembly language (and hence can be run only on DEC-10 and DEC-20
systems). The input syntax closely resembles PDP-11 Macro assembler,
perhaps because CROSS is based upon MACY11, the DEC-10 PDP-11 cross
assembler. Unlike MACY11, however, CROSS can generate output for a wide
variety of micros from basically the same input syntax. But since CROSS
only runs on PDP-10 style processors, these benefits are not widely enjoyed.
To complete the cycle, it would be interesting if someone could write a new
CROSS that accepts PDP-10 Macro-10 for input and produces output in a
variety of formats; then CROSS itself could be CROSS-assembled to form a new
CROSS that could then cross-assemble APPLE.M65 on other machines. But wait,
maybe there's a better way ... ]
------------------------------
Date: 08 Jul 85 AT 15:37:30
From: Ted Medin <MEDIN-T%cc82@NOSC>
Subject: New Apple II Cross Assemblers
I have two Apple II cross assemblers written in C. One can handle the M65
CROSS-format source for Apple II DOS Kermit and the other can handle Apple II
assembler (AP2KER). These assemblers are available in source asis if any one
wants them; I got them from the net and then hacked them. I will mail them in
three parts each as Unix shell archives; cut and execute to produce the files
which include test cases. The test cases are the best documentation on what
the syntax is.
[Ed. - Thanks! The files are available in the Kermit-Tools area on CU20B as
KT:APX*.* (Apple II assembler) and KT:M65*.* (CROSS M65 format), via anonymous
FTP, along with CROSS which is also still there.]
------------------------------
Date: Wed 10 Jul 85 23:12:09-EDT
From: Richard Garland <OC.GARLAND@CU20B.ARPA>
Subject: Re: Kermit on MicroVAX-I
Kermit was the first means we used to load stuff into our microVAX-I last
October under microVMS V1.0. We down-loaded the Stevens Kermit .HEX file
(using raw terminal capture with XON/XOFF flow control), did the same with
the DEHEXIFY Macro source, assembled DEHEXIFY, dehexified KERMIT, and used
it for weeks getting our applictions down from our VMS VAX 780. It was a
mainstay until DECnet over Ethernet got installed. It was hardwired
port-to-port at 9600 bpi. The current version of Kermit works equally well
with microVMS V4.1
Rg
------------------------------
Date: Wed, 10 Jul 85 09:58:22 cdt
From: knutson@ut-ngp.ARPA (Jim Knutson)
Subject: Re: Z100 Communications Program Query
I had thought long and hard about a way to access both ports when I was
getting the Z100 version up. The problem has to do with using the BIOS
routines for character output. These only support one AUX device. To be
able to use the other serial port, a rewrite of the I/O module would need to
be done and some interrupt driven I/O routines would have to be written.
Unfortunately, I don't have the time to do this but it shouldn't be too hard
if you use the IBM-PC I/O module as an example.
Jim Knutson
------------------------------
Date: Fri 12 Jul 85 11:00:54-PDT
From: Wing Lee <WingLee%ECLD@ECLA>
Subject: IBM 3270/PC and Kermit?
Will the IBM-PC kermit work on an IBM 3270/PC using its RS-232 port? We
currently have version 2.26 of IBM-PC Kermit.
wing lee
[Ed. - Anybody have experience with this? My guess is that it would work
just fine, but have never had any reports either way.]
------------------------------
End of Info-Kermit Digest
*************************
-------