informis
/
textfiles
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
textfiles/messages/ALANWESTON/1995/CIS11_04.txt

908 lines
30 KiB
Plaintext
Raw Permalink Normal View History

Initial commit
2021-04-15 11:31:59 -07:00
#: 21229 S1/General Interest
17-Oct-95 04:57:21
Sb: #20940-MVME162
Fm: - Visitor 100442,477
To: Roy D Miller 100354,3307
I also am having problems, not only with the VME162 boards, but with
a 4 way RS232 IP board made by TEWS. I can communicate quite happily if I dont
try to access the RTS line. Once I start doing this, and there is
activity on the other line my software eventually hangs up. I'm at this
moment trying to solve this, and will let you know if I have any success!!!
#: 21228 S1/General Interest
16-Oct-95 21:11:27
Sb: Ram Pak
Fm: Brother Jeremy, CSJW 76477,142
To: All
Dear friends:
I am looking to buy a DISTO 512K Ram Pak. This was a 512k ram disk that
plugged into a multipak. I will buy either in working or reasonably fixable
condition. I would also need any associated software.
With all best wishes,
Brother Jeremy, CSJW
#: 21230 S1/General Interest
17-Oct-95 14:04:07
Sb: #infoxpress
Fm: Jim Vestal 103037,2655
To: All
Hi all. I finally was able to resign up with Compuserve. Now that I can
access it via the internet there is no usuage charges!
Does anyone know how to get InfoXpress to work when connected via telnet (from
my internet shell account)? The only options I see for connecting is direct
dial via the CI$ network or Tymnet/Sprintnet. We do not have local CI$ or
Sprintnet and our locl Tymnet adds a $2 surcharge for CI$ so I refuse to use
it, why pay for an unlimited shell inet account and CI$ too.
Jim
There are 3 Replies.
#: 21231 S1/General Interest
17-Oct-95 19:14:50
Sb: #21230-infoxpress
Fm: David Breeding 72330,2051
To: Jim Vestal 103037,2655 (X)
> Hi all. I finally was able to resign up with Compuserve. Now that I can
> access it via the internet there is no usuage charges!
Hi, Jim. Good to see you back.
> Does anyone know how to get InfoXpress to work when connected via telnet
> (from my internet shell account)? The only options I see for connecting
> is direct dial via the CI$ network or Tymnet/Sprintnet.
Hmmm... An aspect I had not considered - myself not yet having such access :-(
Sounds like another assignment for Bill <G>
You might be able to fake some of the fields in default.ini whereby you could
transfer control over to IX from an already-established connection. However,
apparently IX looks for responses from the modem, so it might take a little
work, but it might be possible.
-- David Breeding --
CompuServe : 72330,2051 Delphi : DBREEDING
*** Composed with InfoXpress/OSK Vr. 1.02 & VED Vr. 2.4.0 ***
#: 21233 S1/General Interest
17-Oct-95 21:02:30
Sb: #21230-#infoxpress
Fm: John Murphy 73077,2305
To: Jim Vestal 103037,2655 (X)
<< Hi all. I finally was able to resign up with Compuserve. Now that I can
access it via the internet there is no usuage charges! >>
Jim, you may be in for a heck of a surprise.
There may not be any SURCHARGES, but the USAGE charges are still in effect,
even if telnet'ing in.
Of course, as long as you keep the usage under 3 (5?) hours per month, then,
yeah, in effect there is no additional charge beyond the monthly membership
fee.
Caveat Emptor...
John
There is 1 Reply.
#: 21234 S1/General Interest
18-Oct-95 02:56:31
Sb: #21233-infoxpress
Fm: Jim Vestal 103037,2655
To: John Murphy 73077,2305 (X)
Actually we get 5 free hours for $9.95 per month, not too bad.
I meant no connect charges, not usuage charges.
On my trial account, my first month is free with 10 FREE hours AND on top of
that I got $35 usuage credits for additional hours. In all, not too bad.
Jim
#: 21235 S1/General Interest
18-Oct-95 04:31:29
Sb: #21230-#infoxpress
Fm: Bill Dickhaus 70325,523
To: Jim Vestal 103037,2655 (X)
Hi Jim, welcome back! Yes, its possible to use a unix shell account with
InfoXpress. I use it during the day with Delphi, to avoid network
charges. I can't guarantee that it will work, though, it depends on a
couple of things working out just right. What version of InfoXpress do
you have? Also, I'm pretty sure that B protocol upload/download won't
work either since I had to use the SevenBit option to get it to work.
You need to set up a special network option block, this is the one I
use:
[CIS Internet] ;Network options for CIS internet connection
NetworkType = alternate ;Network type
Modem = Modem /T0 ;Modem option block name (not path name!)
BaudRate = 9600 ;Baud rate
PhoneNumber = 555-1234 ;Phone number
SevenBit
Macro=cis_internet_macro
Refer to this network option block from the CIS system option block,
rather than the standard one (I keep them both in the ini file with one
or the other commented out). The macro (with a file name of
'cis_internet_macro' just like the option value) looks something like
this:
WAIT login:
SEND userid^M
WAIT word:
SEND password^M
WAIT %
SEND telnet^M
WAIT telnet>
SEND toggle localflow^M
WAIT telnet>
SEND open compuserve.com^M
WAIT Host name:
SEND CISAGREE^M
WAIT OFF):
SEND 9600^M
WAIT User ID:
EXIT
As you can see, these 'macros' are a very simple script languange.
InfoXpress expects to be left at the CIS UserID: prompt, or this won't
work. I can't promise that it will work for you, but I'll do what I can
to help you get it going.
-Bill-
There is 1 Reply.
#: 21238 S1/General Interest
18-Oct-95 20:08:32
Sb: #21235-#infoxpress
Fm: David Breeding 72330,2051
To: Bill Dickhaus 70325,523 (X)
> Hi Jim, welcome back! Yes, its possible to use a unix shell account with
> InfoXpress. I use it during the day with Delphi, to avoid network
> charges.
> As you can see, these 'macros' are a very simple script languange.
You mean we could do this all along? I didn't check the manual, but is all
this in there?
Alright.. maybe what you are gonna have to do is rewrite IX altogether.
You need to incorporate a quiz in the program asking questions from the manual.
Until we can answer 90% of the questions, the program will be inoperative
<G>... And I'm about the worst manual non-reader of the whole group, I'll
bet..
-- David Breeding --
CompuServe : 72330,2051 Delphi : DBREEDING
*** Composed with InfoXpress/OSK Vr. 1.02 & VED Vr. 2.4.0 ***
There is 1 Reply.
#: 21248 S1/General Interest
25-Oct-95 14:27:45
Sb: #21238-#infoxpress
Fm: Bill Dickhaus 70325,523
To: David Breeding 72330,2051 (X)
Um, well, uh....no, its not in the current manual. Mark quit updating the
manual at release 1.1, and I haven't had the time to learn lout, or convert it
to something else. I've got a couple of other things I'm doing right now (you
know, like my real job :-) but I'm hoping to get either 1.3.0 or 2.0 finished
and the manual updated to reflect all the changes. I've also been waiting to
see what CompuServe does about the ascii interface.
-Bill-
There is 1 Reply.
#: 21250 S1/General Interest
26-Oct-95 18:29:55
Sb: #21248-infoxpress
Fm: David Breeding 72330,2051
To: Bill Dickhaus 70325,523 (X)
> Um, well, uh....no, its not in the current manual.
Well.. this makes me feel less foolish, anyway <G>. I have still not studied
your example for the Internet logon yet (not having an Internet acct), but
quickly glancing over it, it does seem to be quite an ingenious implementation.
> Mark quit updating
> the manual at release 1.1, and I haven't had the time to learn lout, or
> convert it to something else. I've got a couple of other things I'm doing
> right now (you know, like my real job :-)
Yeah, aren't stuff like jobs and the like real bothers? Wish we didn't have
them <VBG>
> but I'm hoping to get either
> 1.3.0 or 2.0 finished and the manual updated to reflect all the changes.
> I've also been waiting to see what CompuServe does about the ascii
> interface.
Well, of course I guess we all have our own little "wish lists" for features,
but I, for one, am really thankful we have it in its current version. It is
quite reliable, and I am totally dependent upon it for all my mail and
messaging.
-- David Breeding --
CompuServe : 72330,2051 Delphi : DBREEDING
*** Composed with InfoXpress/OSK Vr. 1.02 & VED Vr. 2.4.0 ***
#: 21241 S1/General Interest
20-Oct-95 21:04:48
Sb: #VGA On The CoCo
Fm: George Hendrickson 71071,2003
To: All
I've been wondering if anyone has put a VGA monitor on there CoCo and if so,
how do you do it? Is it possible?I'm sure that it is....
There is 1 Reply.
#: 21243 S1/General Interest
20-Oct-95 23:16:18
Sb: #21241-VGA On The CoCo
Fm: Rogelio Perea 72056,1204
To: George Hendrickson 71071,2003 (X)
Hello George,
Several articles on old CoCo magazines explained such a setup with a VGA
monitor on a CoCo. It seems today's VGA standard (and since it was implemented)
works on RGB analog video information - contrary to that RGB TTL video from CGA
and EGA, so to use a VGA monitor with a CoCo 3 (which feeds out RGB analog
video) you just need to match up the signal lines between the two connectors.
The "catch" on using VGA is that the VGA monitor MUST be able to synch at 15
+/- MHz (don't recall the exact number) as this is the frequency used by the
CoCo's RGB. Many multi-synch monitors are able to step down this far, still, I
haven't seen such on the newest PC hardware catalogs; there must be some
available new from certain manufacturers.
Perhaps you can get lucky with such a VGA monitor in stores that sell used PC
stuff. This is one of those things I have kept an eye open for, so when I come
across a Multi-Synch monitor capable of beign used with my CoCo 3 I WILL BUY IT
ON THE SPOT!. The quality of the image must be something to see and enjoy.
IF you have some questions about the pinouts on both the VGA and CoCo's RGB
connectors feel free to post again here. Right this minute I do not have my
tech manuals handy and need time to browse the books before posting something
here... ;-)
Rogelio Perea
AZ-TEC Group
#: 21247 S5/OS9 Users Group
24-Oct-95 15:12:05
Sb: #21190-Systemtime via network
Fm: Norbert Wolski 101233,2033
To: M. Raabe 100327,1526 (X)
hello Mr.Raabe,
Thanks a lot for your help !
best regards
N.Wolski
#: 21249 S5/OS9 Users Group
26-Oct-95 17:43:24
Sb: Ethernet printer driver
Fm: RABATE 100333,3004
To: all
Hello,
Does anyone know of an ethernet driver that would run on a 68040 board ?
(freeware, commercial)
Thanks
Nicolas from Paris
#: 21223 S6/Applications
16-Oct-95 19:07:32
Sb: #21221-Terminfo update??
Fm: David Breeding 72330,2051
To: Pete Lyall 76703,4230 (X)
> Your assumptions are correct... tic is 'term-info compiler'.
>
> You should be able to do this on any SysV.4 box (or variants).
>
> If you know anyone who runs LINUX, both tools are available there as well
> (a great, FREE unix for PC's).
I don't know of anyone running LINUX. I have heard some really good reports on
it, but I don't have a PC, myself.
> Should you run out of other options, I have a LINUX box here, but will be
> out of town most of the week. Worst case, you could mail me the file, I'll
> effect the conversions, and return mail you the result.
I'm not sure if it is a case of my terminfo file not being compatible with my
system or not, but it would seem so.. The TERM (vga) screen appears to do well,
but my G-Windows screen has a bit of trouble with the codes..
There is no OSK version of "tic"?
-- David Breeding --
CompuServe : 72330,2051 Delphi : DBREEDING
*** Composed with InfoXpress/OSK Vr. 1.02 & VED Vr. 2.4.0 ***
#: 21224 S6/Applications
16-Oct-95 19:07:39
Sb: #21221-#Terminfo update??
Fm: David Breeding 72330,2051
To: Pete Lyall 76703,4230 (X)
Almost forgot...
> Should you run out of other options, I have a LINUX box here, but will be
> out of town most of the week. Worst case, you could mail me the file, I'll
> effect the conversions, and return mail you the result.
If I should send you the file, would I just need to send you the vt100 section?
This is all I need...
-- David Breeding --
CompuServe : 72330,2051 Delphi : DBREEDING
*** Composed with InfoXpress/OSK Vr. 1.02 & VED Vr. 2.4.0 ***
There is 1 Reply.
#: 21225 S6/Applications
16-Oct-95 20:12:34
Sb: #21224-#Terminfo update??
Fm: John R. Wainwright 72517,676
To: David Breeding 72330,2051 (X)
David,
I got 'captoinfo' and 'tic' by compiling the source in an upload called
'curses.ar' in lib 12. Hope it's still there.
Re Pete's comments on LINUX. It is neat. Got X-windows on the 486 box.
Studying the way the LINUX shell handled the Ghostscript makefile was the key
to finally getting it to compile on the MM/1. Sneaky, huh?
John Wainwright
There is 1 Reply.
#: 21232 S6/Applications
17-Oct-95 19:14:58
Sb: #21225-#Terminfo update??
Fm: David Breeding 72330,2051
To: John R. Wainwright 72517,676 (X)
> I got 'captoinfo' and 'tic' by compiling the source in an upload called
> 'curses.ar' in lib 12. Hope it's still there.
I got that source. I haven't compiled it yet, but I didn't see any reference
to these in it. Do I need something else? Here's the file I got:
file file stored
file name ver file date attr size size
----------------------------- --- -------------- ------ ----- -----
curses.c 0 87/08/09 14:56 ----wr 27301 15850
curses.h 0 87/08/06 22:20 --r-wr 2269 1121
curses.r 0 87/08/09 14:57 ----wr 5093 3582
> Re Pete's comments on LINUX. It is neat. Got X-windows on the 486 box.
I've seen references to it in a PC-related newsgroup my local BBS downloads. It
really sounds like a good setup.. Maybe I will have to get a PC after all...
> Studying the way the LINUX shell handled the Ghostscript makefile was the
> key to finally getting it to compile on the MM/1. Sneaky, huh?
Very.. <G>
-- David Breeding --
CompuServe : 72330,2051 Delphi : DBREEDING
*** Composed with InfoXpress/OSK Vr. 1.02 & VED Vr. 2.4.0 ***
There is 1 Reply.
#: 21236 S6/Applications
18-Oct-95 18:22:11
Sb: #21232-#Terminfo update??
Fm: John R. Wainwright 72517,676
To: David Breeding 72330,2051 (X)
> > 'curses.ar' in lib 12. Hope it's
> still there.
>
> I got that source. I haven't compiled
> it yet, but I didn't see any reference
> to these in it. Do I need something
> else? Here's the file I got:
OOps, checked my list again. There are TWO files in DL-12 called "curses.ar".
One with length 20841 by 76625,2440 is probably the one you have
The other, with length 264742 by 73270,3124 is the one I got the captoinfo and
tic source from. There is a lot of other stuff, including a game called
"MILLE" in there too.
John
There are 2 Replies.
#: 21237 S6/Applications
18-Oct-95 20:08:25
Sb: #21236-Terminfo update??
Fm: David Breeding 72330,2051
To: John R. Wainwright 72517,676 (X)
> > I got that source. I haven't compiled
> > it yet, but I didn't see any reference
> > to these in it. Do I need something
> > else? Here's the file I got:
>
> OOps, checked my list again. There are TWO files in DL-12 called
> "curses.ar". One with length 20841 by 76625,2440 is probably the one you
> have
It sure is.. so that's why I am missing out. Guess I'll have to go get the
other one..
> The other, with length 264742 by 73270,3124 is the one I got the captoinfo
> and tic source from.
Thanks, John. I think maybe I can get it up and running now..
-- David Breeding --
CompuServe : 72330,2051 Delphi : DBREEDING
*** Composed with InfoXpress/OSK Vr. 1.02 & VED Vr. 2.4.0 ***
#: 21239 S6/Applications
18-Oct-95 20:08:38
Sb: #21236-Terminfo update??
Fm: David Breeding 72330,2051
To: John R. Wainwright 72517,676 (X)
> > > 'curses.ar' in lib 12. Hope it's
> > still there.
> The other, with length 264742 by 73270,3124 is the one I got the captoinfo
> and tic source from.
It doesn't seem to be here. The only one I could find was the one I have.
Don't they keep an archive of old files here? Maybe I can get them to repost
it.
-- David Breeding --
CompuServe : 72330,2051 Delphi : DBREEDING
*** Composed with InfoXpress/OSK Vr. 1.02 & VED Vr. 2.4.0 ***
#: 21226 S10/OS9/6809 (CoCo)
16-Oct-95 21:08:43
Sb: Ram Pak
Fm: Brother Jeremy, CSJW 76477,142
To: all
Dear friends:
I am looking to buy a DISTO 512K Ram Pak. This was a 512k ram disk that
plugged into a multipak. I will buy either in working or reasonably fixable
condition. I would also need any associated software.
With all best wishes,
Brother Jeremy, CSJW
#: 21242 S10/OS9/6809 (CoCo)
20-Oct-95 21:08:14
Sb: IDE Drives on the CoCo
Fm: George Hendrickson 71071,2003
To: All
Is it possible to connect an IDE drive to the CoCo using a HyperIO interface
card and an 8 bit IDE controller?
#: 21222 S12/OS9/68000 (OSK)
16-Oct-95 02:28:19
Sb: #21216-#OS-9 on MVME162LX
Fm: Jost Eberbach 73502,2041
To: Manfred Berger 76570,2207 (X)
Hello Manfred,
regarding the off-board memory Ole Hansen already gave you some good hints. The
VMEchip needs some registers initialized, before you can access the VMEbus
safely. The OS-9 debugger does that automatically for you, but apparently not
the Mototrola Rombug.
For the serial ports, I also tried to use higher baud rates. I found out, that
with OS-9 version 2.4 the device driver is not able to use a higher baud rate
than 19200, although the chips are specified to go up to a speed of 38.4kbaud.
I think with OS-9 3.0 this is fixed, but I'm not sure, I never tried it. Ole is
right, that it doesn't make much sense to use more than 9600 baud on the
console (/term) port, because the debugger uses that baudrate as default, and
it's not easy to change that. But it makes some sense to use a higher baud rate
on serial port 2 (/t1), especially if you use it to download files.
Regards,
Jost
There is 1 Reply.
#: 21244 S12/OS9/68000 (OSK)
21-Oct-95 14:40:03
Sb: #21222-#OS-9 on MVME162LX
Fm: Manfred Berger 76570,2207
To: Jost Eberbach 73502,2041 (X)
Hello Jost;
I have tried tmode and I wrote my own program to set the baud rate to 38400 but
OS-9 3.0 doesn't allow me to do so. Since I am cross developing it would
really nice to speed up the download time and I'm also planning to do real-time
data acquisition and transfer the data over the RS-232 port. Since I only have
the BSP product I don't have source code for the serial driver so I can't
change it and I really don't want to write my own driver.
Manfred
There is 1 Reply.
#: 21245 S12/OS9/68000 (OSK)
24-Oct-95 11:41:56
Sb: #21244-OS-9 on MVME162LX
Fm: Jost Eberbach 73502,2041
To: Manfred Berger 76570,2207 (X)
Manfred,
>>I have tried tmode and I wrote my own program to set the baud rate to 38400
but OS-9 3.0 doesn't allow me to do so. Since I am cross developing it would
really nice to speed up the download time and I'm also planning to do real-time
data acquisition and transfer the data over the RS-232 port. Since I only have
the BSP product I don't have source code for the serial driver so I can't
change it and I really don't want to write my own driver.<<
Weel, I guess you have to accept what MW did, if you don't want to write your
own driver. Have you considered using Ethernet (TCP/IP) for transfering the
data? Another option is to purchase a third party serial board for higher baud
rates, incl OS-9 driver, and use that. But make sure you get a driver for OS-9
3.0, the ones for earlier version might not work...
I know of an english company who have such boards and drivers. If you like, I
can give you their address, I don't have it right now, but I can find it
somewhere.
Regards,
Jost
#: 21227 S12/OS9/68000 (OSK)
16-Oct-95 21:10:39
Sb: Ram Card
Fm: Brother Jeremy, CSJW 76477,142
To: All
Dear friends:
I am looking to buy a DISTO 512K Ram Pak. This was a 512k ram disk that
plugged into a multipak. I will buy either in working or reasonably fixable
condition. I would also need any associated software.
With all best wishes,
Brother Jeremy, CSJW
#: 21240 S12/OS9/68000 (OSK)
20-Oct-95 11:24:50
Sb: #OS9 3.0 crashes if ...
Fm: Rainer Thieringer 100544,1230
To: All
Hello,
the last few weeks I've had serious trouble with OS9/68000. I hope this message
will prevent anyone out there from spending lonely nights face to face with a
crashed system.
Scenario 1: A board in my system produces asynchronous interrupts.
Unfortunately the interrupt level switched to the VME bus had intermediate
states (e.g. IRQ 6 was for a few us an IRQ 2). This should be reported as
SPURIOUS IRQ followed by a system reset if not handled by a user trap (or 3.0's
SpurIRQ flag). What I've had was a total CRASH !!! Not even ROMBUG could help.
I (and my hardware vendor) suppose that this SPURIOUS IRQ message will not
appear, if immediately after the first (bad) IRQ another (good) IRQ happens.
This crashes OS-9 completely.
Workaround: a) I set the SpurIRQ-flag in the init modules COMPAT Byte to ignore
Spurious IRQs. b) I jumpered the board with IRQ 2 so that no intermediate IRQ
states could happen.
Scenario 2: Together with that SpurIRQ flag I set the PrivAlm flag which should
keep OS-9 from accidentialy deleting foreign alarms. After I did so I've had
system crashes (w/o any notification) every couple of minutes. After resetting
the flag everything worked fine again. As I've heard on the SYSTEMS this is an
know bug. Microware offers an update.
BUT ONCE AGAIN MICROWARE DID NOT INFORM US DEVELOPERS ABOUT A KNOWN BUG!
I hope this was the last serious surprise with 3.0x (knock on wood).
There is 1 Reply.
#: 21246 S12/OS9/68000 (OSK)
24-Oct-95 12:49:31
Sb: #21240-#OS9 3.0 crashes if ...
Fm: Ian J Shearer 100410,2733
To: Rainer Thieringer 100544,1230 (X)
Rainer,
>>I (and my hardware vendor) suppose that this SPURIOUS IRQ message will not
appear, if immediately after the first (bad) IRQ another (good) IRQ happens.
This crashes OS-9 completely.<<
This isn't really down to OS-9, is it?. If a 680x0 processor gets an exception
during an exception-handling routine, it resets. It looks as if your problem
was poor hardware rather than poor OS.
>>I set the PrivAlm flag which should keep OS-9 from accidentialy deleting
foreign alarms<<
The PrivAlm seems to be something of a workaround for slightly flaky alarms.
Doesn't it only work to protect other UserID's (no use for tasks with a common
parent)? You can use semaphores to guard against accidental deletes, which
avoids the M'ware bug.
>>BUT ONCE AGAIN MICROWARE DID NOT INFORM US DEVELOPERS ABOUT A KNOWN BUG!<<
Microware have NEVER been good at advertising their faults.
>>I hope this was the last serious surprise with 3.0x (knock on wood).<<
Now that would be surprising!!
Ian
-Ian J Shearer
There is 1 Reply.
#: 21255 S12/OS9/68000 (OSK)
02-Nov-95 03:57:18
Sb: #21246-#OS9 3.0 crashes if ...
Fm: Rainer Thieringer 100544,1230
To: Ian J Shearer 100410,2733 (X)
Hello Ian,
nice to get some response.
> This isn't really down to OS-9, is it?. If a 680x0 processor gets an
>exception during an exception-handling routine, it resets. It looks as if
your
>problem was poor hardware rather than poor OS.
> OK Ian, I must admit, that I am not very familiar with the 680x0 processors.
Fact is, that I haven't ever had problems with OS9 2.4. That doesn't mean that
I haven't ever got spurious IRQs! Boy, I really got a hell lot of them, but OS9
2.4 told me that it has caught one before the system resets. So probably you
are right, but if so I hope that my report gives someone having these
mysterious crashes a hint.
> The PrivAlm seems to be something of a workaround for slightly flaky alarms.
>Doesn't it only work to protect other UserID's (no use for tasks with a
common
>parent)? You can use semaphores to guard against accidental deletes, which
>avoids the M'ware bug.
> In OS-9 Technical Manual pg. 2-29 the description to M$Compat says: "Only
the *process* that created an alarm can delete it". If you know more about that
flag, please let me know.
- Rainer Thieringer
HAAS LASER GmbH
Germany
There is 1 Reply.
#: 21256 S12/OS9/68000 (OSK)
03-Nov-95 08:57:05
Sb: #21255-OS9 3.0 crashes if ...
Fm: Ian J Shearer 100410,2733
To: Rainer Thieringer 100544,1230 (X)
Rainer,
>> The PrivAlm seems to be something of a workaround for slightly flaky
alarms.
>>Doesn't it only work to protect other UserID's (no use for tasks with a
common
>>parent)? You can use semaphores to guard against accidental deletes, which
>>avoids the M'ware bug.
> In OS-9 Technical Manual pg. 2-29 the description to M$Compat says: "Only
the *process* that created an alarm can delete it". If you know more about that
flag, please let me know.
You're right, that's what the manual says. I was basing my comments on
information from Microware Technical Support in the UK. To quote . . .
"On Ver3.0 OS-9 this may be partially prevented by setting a field in the init
module, which causes a check such that when an alarm is deleted, it must be in
the same user/group."
I queried the statement at the time, and was told (verbally) that sibling
processes sharing the same user/group ID could still accidentally delete each
others alarms. Since this didn't solve my problem, I used a workaround rather
than M$Compat.
Not a complete answer, I know, but the best I can do for now. Hope it helps.
-Ian J Shearer, Onyx Systems Ltd.
#: 21252 S12/OS9/68000 (OSK)
30-Oct-95 22:50:36
Sb: #SCF hardware handshake
Fm: Doug Bailey 73612,2146
To: sysop (X)
I am writing an OS-9 application to a device that uses RTS/CTS flow control. I
need to enable my SCF device to handle this flow control but do not know how.
I have looked at changing the device descriptor but nothing seems obvious
there. I see that the SCF device has a static structure that signals when to
enable the RTS/CTS but don't know how to access it?
Anybody with any advice?
Thanks for the help!!
Doug Bailey
There are 2 Replies.
#: 21253 S12/OS9/68000 (OSK)
31-Oct-95 05:35:40
Sb: #21252-#SCF hardware handshake
Fm: Marc Tritschler 100344,2706
To: Doug Bailey 73612,2146 (X)
The ability of your system to work with CTS/RTS flow control depends on two
things - the ability of your actual serial communications device to handle flow
control, and the device driver itself having the ability to configure the
device to do so.
It is safe to assume that your serial device has the ability to handle flow
control, but the device driver is another question. The device driver will
have, most likely, been supplied to you by the company you have bought your
hardware from, and so I would recommend that you speak to them to get an
accurate answer.
I will give you an example of my SCF device driver experiences. In the SCF
device drivers that I have, CTS / RTS flow control is enabled by using the
xmode utility to set the XON and XOFF properties to certain values not normally
used for XON and XOFF e.g. both to 0. The device driver, when it picks this up,
puts the device into CTS / RTS flow control mode.
Unfortunately, however, the manufacturers definition of flow control was not
the same as mine - they asserted RTS permanently (whether or not data was ready
to be transmitted) and waited for CTS to be asserted before transmitting data,
whereas I required to assert RTS only when I had data available to send, and
wait for CTS to be asserted before sending it, then de-assert RTS after the
data was sent. In order to achieve this I had to modify the device driver
itself, which required fairly in-depth knowledge about the device.
Obviously, if you dont have this knowledge, and need to do a similar
modification, it is better to get the manufacturer / device driver supplier to
do it for you.
I hope this puts things into perspective.
Cheers,
Marc.
There is 1 Reply.
#: 21254 S12/OS9/68000 (OSK)
31-Oct-95 21:46:09
Sb: #21253-SCF hardware handshake
Fm: Doug Bailey 73612,2146
To: Marc Tritschler 100344,2706 (X)
Thanks for the info Marc.
My situation is that I'm running on a MVME167 box wwith a 68681 UART device.
However, the eventual target is a custom box again using a 68681 UART. I have
source code for the driver and it has a flag stored in the SCF static data
field that allows me to activate the hardware flow control. It seems I need to
get to this flag before driver initialization.
I guess my best bet is as you recommended and talk to the driver's author.
Thanks again,
Doug Bailey
#: 21257 S12/OS9/68000 (OSK)
03-Nov-95 09:29:53
Sb: #21252-#SCF hardware handshake
Fm: Jost Eberbach 73502,2041
To: Doug Bailey 73612,2146 (X)
Doug,
there is an OS-9 system call that you could use to enable RTS/CTS flow control.
The name of the call is I$SetStt. Among others, there are function codes
available to disable and enable the RTS hardware handshake lines. From C, you
can use that call by calling _os_ss_dsrts(path) to disable or
_os_ss_enrts(path) to enable RTS (see your C-library manual). Of course, this
requires that your driver as well as your hardware support RTS. You could
simply try it, or consult your hardware manufacturer, in case your manuals
don't tell you enough about RTS and CTS (mine don't, I'm using an ELTEC board
right now). They should be able to help you.
Regards,
Jost
There is 1 Reply.
#: 21258 S12/OS9/68000 (OSK)
03-Nov-95 12:25:20
Sb: #21257-SCF hardware handshake
Fm: Doug Bailey 73612,2146
To: Jost Eberbach 73502,2041 (X)
Thanks for the info. I'll try it out.
Doug Bailey
#: 21259 S12/OS9/68000 (OSK)
04-Nov-95 14:10:29
Sb: Windio #56/Screen Saver
Fm: Boisy G. Pitre 74464,3005
To: All
Windio #56 is now available in section 12 of the library.
Press <CR> !>