home *** CD-ROM | disk | FTP | other *** search
- Date: 28 Feb 1989 1513-EST
- From: EIBEN@tops20.dec.com
- To: Frank da Cruz <fdc@cunixc.cc.columbia.edu>
- Subject: Rainbow Kermit
-
- Here a 'solicited' response from a Rainbow Kermit user.
-
- Rgds,
- bernie.
-
- <<< HUMAN::DISK$HUMAN_WRKD:[NOTES$LIBRARY]KERMIT.NOTE;1 >>>
- -< KERMIT reopen for business >-
- ================================================================================
- Note 602.1 Modified MS-Kermit for Rainbow available 1 of 1
- IOENG::METZ "Better Latent than not at all" 310 lines 27-FEB-1989 18:16
- -< Comm Bug and Feedback >-
- --------------------------------------------------------------------------------
-
- First of all Thanks!!!
-
- It's great to have the latest and greatest Kermit for
- the Rainbow. I use Kermit all the time and just found out about
- the Script feature (which could save me about an hour a day).
-
- You asked for some feedback on using 2.32/A so here it is.
-
-
- Used Kermit in the following Environment:
-
- a) Printer printing while Kermit was running a 'dial' script -
- works.
-
- b) Changing SPEED interactively in a 'dial' script file - works.
-
- c) Kermit 'push' and SEDT cycles - works.
-
- d) Logging sessions - works.
-
- e) SET key VT220 definitions - works.
-
- f) Interactive sessions on VMS (used remote server commands,
- connect and file transfers) - works.
-
-
- -------------------------------------------------------------------------
- The following identifies some comments/problems with Kermit-MS Version
- 2.32/A dated 21-Jan-1989 when running on a Rainbow 100+ under MSDOS V2.11
- -------------------------------------------------------------------------
-
- SHOW MODEM
- SHOW COMMUNICATIONS
-
- Regardless of the modem status or the communications set-up
- the following status is always displayed:
-
- DSR is off
- CD is on
- CTS is off
-
- Modem cable is a BCC22-B (14 wire) connected to a Scholar modem.
-
- Because of the false status above the following commands will also
- not function:
-
- WAIT [{number, hh:mm:ss}] [\CD] [\CTS] [\DSR] - always times out.
-
- HANGUP - does not drop DTR/RTS.
-
-
- [ NOTE: I Debugged all above modem status problems - See end of note for
- comments].
-
-
- * Standard Operating Procedure, Bug or feature?
-
- 1) 'SHOW?' will only display 'SHOW' as an option where as 'SHOW ?'
- will display all 'SHOW Options'. (Does the same for all, eg SET
- etc.)
-
-
- 2) Using a macro embedded in text failed sometimes (dependent
- on text/definition?) by flaging the text after the macro as
- '?text'. I had to split the ECHO line in order for it to work.
- [I don't remember what the actual line was which failed but
- I do remember it was long. If I find it I'll include it here].
-
- Example only (This does work):
-
- ECHO \07\13 \%m test\07 - This or
- ?test
-
- ECHO \07\13\%mtest\07 - that Failed but...
- ?test
-
-
- ECHO \07\13\%m - this worked
- ECHO test\07 - when I split the line.
-
-
-
- 3) When using a Script to log into a VMS system and 'ECHO INPUT
- OFF' is set. The terminal identifier sequence from the VAX
- identifies the terminal as UNKNOWN. [This could be an XOFF
- problem as I noticed that there is a backlog of output (to
- the rainbow) starting with the UNKNOWN terminal id once the
- script enters CONNECT and I hit return].
-
-
-
- * Known Kermit-MS 2.32/A problems noted elsewhere (sorry I didn't give
- credit to those who documented but I did run into the problem).
-
- 1) Labels require a appended space and comment in order to be
- 'found'.
-
- Instead of :LABEL
-
- Use :LABEL ;232/A fix for using labels
-
-
-
- * Operational Notes/Wishes:
-
- 1) Use double ^C to exit Script file execution. (Would be
- nice to have a 'ON CONTROL_C GOTO LABEL' to abort modem).
-
-
- 2) When using a script file to specify a .com file on a VMS host
- be careful how you specify the '@file'.
-
- Instead of OUTPUT @file.com\13 ;which uses local @file for output
-
- Use OUTPUT \13@file.com\13 ;this will run the VMS @file
-
- ; KERMIT.EXE Version: 2.32/A 21 Jan 1989 - Modem Control Patches
- ;
- ; To fix the modem handshaking problems (I needed to have it work quickly) I
- ; got a copy of MSXRB1.ASM on CSCMAS which I used to help debug and patch
- ; (just to many files to downline load slowly... and reassemble). All modem
- ; handshaking now works (DTR/RTS/SRTS/DSR/CTS/RLSD). Here's how and what I
- ; changed (not in order of 'found' as you can tell by the comments):
- ;
- ; These addresses reflect DEBUGs address which is (+100) of the PATCH
- ; program address I used.
-
- >ren kermit.exe kermit.tmp
- >debug kermit.tmp
-
- ; This fixes the problem where all modem status bits were
- ; sampled inverted (RB does this) and therefore was shown
- ; incorrectly under SHOW MODEM or SHOW COMM.
- ;
- ; Was Now
- ; ----------------- ------------------
- ; MOV AL,0 MOV DX,mnctrl
- ; MOV DX,mnctrl IN AL,DX
- ; IN AL,DX NOT AL ;invert RB modem bits
- ;
- ; Note: I left out the MOV AL,0 instruction as I assumed
- ; that the IN AL,DX over-wrote the AL anyhow.
- ;
- ; WAS: B0 00 BA 02 00 EC
- ; NOW: BA 02 00 EC F6 D0
- ;
- E 87BF
- BA 02 00 EC F6 D0
-
-
- ; This fixes the problem where RSLD (Carrier Detect) never
- ; turned on under SHOW MODEM or SHOW COMM and WAIT n \CD
- ; never worked (it was only syntax).
- ;
- ; Was Now
- ; ------------------- --------------------
- ; getmod2: test al,10 getmod2: test al,10h
- ;
- ; was: 0A now: 10
- E 87DD
- 10
-
- ; This fixes the problem where if carrier detect was 'On'
- ; that the DSR and CTS status was sometimes corrupted
- ; under SHOW MODEM or SHOW COMM.
- ;
- ; Was Now
- ; ----------------- ------------------
- ; MOV mdmhand,80h OR mdmhand,80h
- ;
- ; was: C6 06 now: 80 0E
- E 87E0
- 80 0E
- ; Page 2 of 2
-
- ; This corrects the CONNECT so that only DTR/RTS/SRTS turns
- ; on instead of all bits and some LEDs on the back of the
- ; Rainbow. (The LEDS would be a nice feature for standard
- ; programming to display modem status). Functionally the
- ; original code worked ok.
- ;
- ; Was Now
- ; ----------- ----------
- ; MOV AL,030h MOV AL,F1h
- ;
- ;
- E 8B43
- F1 ; from 30 to F1
-
-
- ; This fixes the command 'HANGUP; problem where DTR/RTS did not
- ; drop and the modem stayed connected.
- ;
- ; Was Now
- ; ----------- ----------
- ; MOV AL,00h MOV AL,FFh
- ;
- E 8CBF ;
- FF ; from 00 to FF
-
-
- ; The following changes update a copy of the kermit
- ; version to reflect the above changes however I'm not
- ; sure where this is used or displayed.
- ;
- E 9EAF ; Change 2'1' to 2'5' (date)
- 35
- E 9EB1 ; Change 'Jan' to 'Feb'
- 46 65 62
- E 9EAC ; change '/A ' to '/AB'
- 42
-
- ; The following changes update the kermit welcome and VERSION
- ; command output display to reflect the above patches.
- ;
- E AC85 ; Change 2'1' to 2'5' (date)
- 35
- E AC87 ; Change 'Jan' to 'Feb'
- 46 65 62
- E AC82 ; change '/A ' to '/AB'
- 42
-
- W ; writes patched file
- Q ; Quit debug
-
- >ren kermit.tmp kermit.exe
-
- ;
- ; End of patching
- ;
-
- And for reference:
-
- RBCOMM.HLP Updated: 25-Feb-1989
- ------------------------------------
-
- The Rainbow CP/M technical documentation appears to be incorrect with respect
- to port 02h active/asserted conditions. This is an attempt to clarify the
- 'tested' Comm bit functionality and levels.
-
-
- Rainbow COMM port 02H functions:
- --------------------------------
-
- * Port 02H - Read definitions (Inputs)
-
- 7 6 5 4 3 2 1 0
- --- --- --- --- --- --- --- ---
- Int Int MHFU RLSD CTS DSR SI/ RI
- Z80 L 88 L Ebl L (CD) SRLSD
-
- 80h 40h 20h 10h 08h 04h 02h 01h
- ?=On ?=On ?=On 0=On 0=On 0=On 0=On 0=On
- ?=Off ?=Off ?=Off 1=Off 1=Off 1=Off 1=Off 1=Off
-
- Typical: 1 1 0 x x x x x
-
- Cable: | pin 8 pin 5 pin 6 pin 12 pin 22 |
- RB output: | <--- On = +6V, Off = -6V ---->|
-
-
-
-
- * Port 02H - Write definitions (outputs)
-
- 7 6 5 4 3 2 1 0
- --- --- --- --- --- --- --- ---
- LED5 LED6 LED4 LED7 RTS DTR SRTS Speed
- Select
-
- Cable: pin 4 pin 20 pin 19 pin 23
-
- 80h 40h 20h 10h 08h 04h 02h 01h
- 0=On 0=On 0=On 0=On 0=On 0=On 0=On 0=On (Low?)
- 1=Off 1=Off 1=Off 1=Off 1=Off 1=Off 1=Off 1=Off(High?)
-
-
- *Note: the above LED bits do not match the Status led positions.
-
- View: Back of Rainbow
- --------------------- Status Leds
-
- ------- ------ ------ 1 2 3 4 5 6 7
- \ / \ / \ / o o o o o o o
- ----- ---- ----
-
-
- Rainbow Comm port EIA levels
- ----------------------------
- Power up or writing a 1 (off) shows all output pins at -6V (marked/off).
- Writing a 0 (on) sets output levels to +6V (Spaced/On). All bits function
- correctly. (for reference, Xmit Data pin 2 is a -6V or marked when idle).
- [Scholar modem runs -12V to +12V].
-
-
- Output level specifications:
-
- EIA Data Control Typical
- === ==== ======= =======
-
- +15V ----------------- (+12V)
- Space On
- + 3V ----------------- (+5V)
- ^
- 0V - Invalid Area -
- v
- - 3V ----------------- (-5V)
- Mark Off
- -15V ----------------- (-12V)
-
- --------
-
- Date: Tue, 28 Feb 89 15:33 MDT
- From: Joe Doupnik <JRD@cc.usu.edu>
- Subject: Rainbow updates
- To: eiben@tops20.dec.com, fdc@cunixc.cc.columbia.edu
- X-Vms-To: IN%"eiben@tops20.dec.com", IN%"fdc@cunixc.cc.columbia.edu",JRD
-
- Bernie and Frank,
-
- What a thorough test of Rainbow Kermit. Thanks!
-
- SHOW? is a feature since it implies help on the single keyword
- to-date, SHOW, whereas SHOW ? asks about an empty keyword (whatever can
- follow the SHOW keyword). Abbreviations are shown by the word? form.
-
- Modem bits. Much thanks for this stuff. My DEC Rainbow book is
- clearly full of errors and I could not figure out the proper polarity of
- signals at the programmer's end of things (not to mention a blunder or two).
-
- Script ECHO. I rather suspect the \%m macro contained an explicit
- (bare) carriage return vs one of the \13 things. That would terminate the
- Echo command, leaving the rest for the parser to clean up as noted.
-
- The :LABEL<optional space> thing should not have needed the trailing
- space (an old problem since cured, unless the MSVRB1.BOO file was not the
- most recent version).
-
- Terminal ident response. That ought to work but have not tested it
- here during the one brief chance to use a Rainbow. I'll check the code
- again anyway.
-
- Things that worked. Will wonders never cease?
- Thanks again Bernie,
- Joe D.
-
- ------------------------------
-
-