11 Dec
2000
11 Dec
'00
3:37 a.m.
Apparently, this didnt make it here :-(
Ram
----- Original Message -----
From: <rmeenaks(a)olf.com>
Newsgroups: comp.sys.transputer,comp.os.parix
To: <classiccmp(a)classiccmp.org>
Sent: Sunday, December 10, 2000 9:14 AM
Subject: HELP: BBK-S4 and Solaris 2.6
Hi,
I am trying to get a BBK-S4 SBus-based transputer card working on a SUN
IPX workstation running Solaris 2.6. I have the Basic Transputer
Support Software (BTSS 3.1) which has support for Solaris. However, I
do not have any documentation. The BTSS requires the creation
of "/dev/sbus?" in order for it to communicate with the transputer. I
do not have any /dev/sbus[0-3] devices on my system. I tried to
symbolically link one of the sbus psuedo devices that is found
in /devices to /dev/sbus[0-3], but that doesnt seem to work. How do
you create the /dev/sbus[0-3] devices? I am trying to get Parix 1.2
running on a T8-based Xplorer. This is the 1st step in many to come to
get this system up and running. Any help is much appreciated...
Oh, before anyone asks, I booted using "boot -r" to reconfigure the
system, but that didnt work either...
Ram
Sent via Deja.com http://www.deja.com/
Before you buy.
11 Dec
11 Dec
7:07 a.m.
> Hi,
>
> I am trying to get a BBK-S4 SBus-based transputer card working on a SUN
> IPX workstation running Solaris 2.6. I have the Basic Transputer
> Support Software (BTSS 3.1) which has support for Solaris. However, I
> do not have any documentation. The BTSS requires the creation
> of "/dev/sbus?" in order for it to communicate with the transputer. I
> do not have any /dev/sbus[0-3] devices on my system. I tried to
> symbolically link one of the sbus psuedo devices that is found
> in /devices to /dev/sbus[0-3], but that doesnt seem to work. How do
> you create the /dev/sbus[0-3] devices? I am trying to get Parix 1.2
> running on a T8-based Xplorer. This is the 1st step in many to come to
> get this system up and running. Any help is much appreciated...
>
> Oh, before anyone asks, I booted using "boot -r" to reconfigure the
> system, but that didnt work either...
>
> Ram
>
>
A symbolic link won't work, since its not the contents of the
file that are important. The first place to look is the
MAKEDEV script in the /dev directory. This should show you
how to make the /dev/sbus files.
--
Dr. Mark Green mark(a)cs.ualberta.ca
McCalla Professor (780) 492-4584
Department of Computing Science (780) 492-1071 (FAX)
University of Alberta, Edmonton, Alberta, T6G 2H1, Canada
7:50 a.m.
Thanks for the response, but there is no MAKEDEV on Solaris. You are
supposed to run "boot -r" to reconfigure /dev and /devices and it should
automatically appear. But it didnt happen. :-(
Ram
Mark Green wrote:
> Hi,
>
> I am trying to get a BBK-S4 SBus-based transputer card working on a SUN
> IPX workstation running Solaris 2.6. I have the Basic Transputer
> Support Software (BTSS 3.1) which has support for Solaris. However, I
> do not have any documentation. The BTSS requires the creation
> of "/dev/sbus?" in order for it to communicate with the transputer. I
> do not have any /dev/sbus[0-3] devices on my system. I tried to
> symbolically link one of the sbus psuedo devices that is found
> in /devices to /dev/sbus[0-3], but that doesnt seem to work. How do
> you create the /dev/sbus[0-3] devices? I am trying to get Parix 1.2
> running on a T8-based Xplorer. This is the 1st step in many to come to
> get this system up and running. Any help is much appreciated...
>
> Oh, before anyone asks, I booted using "boot -r" to reconfigure the
> system, but that didnt work either...
>
> Ram
>
>
A symbolic link won't work, since its not the contents of the
file that are important. The first place to look is the
MAKEDEV script in the /dev directory. This should show you
how to make the /dev/sbus files.
8:17 a.m.
Ram Meenakshisundaram wrote:
Thanks for the response, but there is no MAKEDEV on Solaris.
That's correct. It's mknod, although it's use is somethat deprecated in
favor of creating symbolic links to the physical name of the device in
/devices.
You are
supposed to run "boot -r" to reconfigure /dev and /devices and it should
automatically appear. But it didnt happen. :-(
Uh, boot -r isn't a configuration panacea. It causes add_drv and drvconfig
to be invoked which aren't going to deal well with random third-party devices
stuffed onto the bus.
I'd hit /var/adm/messages to make sure that the thing is being sized (Solaris
always sizes the bus, regardless of the -r argument to boot), and if so
go edit /etc/path_to_inst by hand and either use mknod or create your own
symlinks to finish the job -- unless, of course, the device isn't being
sized, in which case you've got a hardware issue to chase after first.
--
Chris Kennedy
chris(a)mainecoon.com
http://www.mainecoon.com
PGP fingerprint: 4E99 10B6 7253 B048 6685 6CBC 55E1 20A3 108D AB97
10:38 a.m.
New subject: Glass media hard drives
If glass is a slow moving liquid as demonstrated by 50 year old running
window glass, how do drive makers expect thier media not to 'dish out'
from the centrifugal force exerted on them?
I'm sure glass media is not all that uncommon, but the only drives I KNOW
use it are the 2.5" Toshiba drives used in notebooks. You have to use a
lot of force, but if you whip one at a concrete floor, the platters will
break into zillions of pieces.
Glass, and the fact that these particular drives are none too reliable
made me wonder.
--
-----------------------------------------------------------
Jeffrey S. Worley
President
Complete Computer Services, Inc.
30 Greenwood Rd.
Asheville, NC 28803
828-277-5959
Visit our website at HTTP://www.Real-Techs.com
THETechnoid(a)home.com
-----------------------------------------------------------
10:37 a.m.
New subject: Glass media hard drives
On Mon, Dec 11, 2000 at 04:38:31PM -0500, THETechnoid(a)home.com (THETechnoid(a)home.com)
wrote:
If glass is a slow moving liquid as demonstrated by 50
year old running
window glass
It ain't. See <URL:http://www.glassnotes.com/WindowPanes.html>.
-Rich
--
------------------------------ Rich Lafferty ---------------------------
Sysadmin/Programmer, Instructional and Information Technology Services
Concordia University, Montreal, QC (514) 848-7625
------------------------- rich(a)alcor.concordia.ca ----------------------
11:49 a.m.
New subject: Glass media hard drives
If glass is a
slow moving liquid as demonstrated by 50 year old running
window glass
It ain't. See <URL:http://www.glassnotes.com/WindowPanes.html>. 
11:35 a.m.
New subject: Glass media hard drives
If glass is a slow moving liquid as demonstrated by 50
year old running
window glass, how do drive makers expect thier media not to 'dish out'
from the centrifugal force exerted on them?
http://math.ucr.edu/home/baez/physics/glass.html
--
Roger Ivie
rivie(a)teraglobal.com
Not speaking for TeraGlobal Communications Corporation
11:43 a.m.
New subject: Glass media hard drives
THETechnoid(a)home.com wrote:
If glass is a slow moving liquid as demonstrated by 50
year old running
window glass, how do drive makers expect thier media not to 'dish out'
from the centrifugal force exerted on them?
Hi
There is no evidence that 50 year old glass runs.
This was an erroneous conclusion from looking at old
stained glass. This old glass was not made like float
glass of today and almost always had some wedge. The
glass was placed in the came with the fat end down
to help in shedding water.
There are both lenses and mirrors made over 100 years
ago that are holding an optical shape that would be
impossible to keep with even the slightest running or
slipping ( less then 2X10^-9 inches ). I don't find it
conceivable that adding another 200 years could make a
change on the order of tenths of an inch, as in the case
of the stained glass. Many of the finest refractor lenses
made and still in use were made in the last 50 to 100 years ago.
These could not be useful at all with the suggested
running stated.
This is an argument the comes up every now and again
that just doesn't make sense. The only thing that 50 year
old glass does is what is called surface chemical tempering.
This is why it is hard to cut old glass. This is caused
by chemicals diffusing in the glass surface ( a method often used
to temper glass for special purposes to day ).
While it is true that glass is still semi liquid, like the
rings of Saturn that are made of ice and will sublime over
time, the rate of sagging of glass at normal temperature is
no relevant on human time scales.
Dwight
8:14 a.m.
On Mon, 11 Dec 2000, Ram Meenakshisundaram wrote:
> IPX workstation running Solaris 2.6. I have the
Basic Transputer
> Support Software (BTSS 3.1) which has support for Solaris. However, I
> do not have any documentation. The BTSS requires the creation
> of "/dev/sbus?" in order for it to communicate with the transputer. I
> do not have any /dev/sbus[0-3] devices on my system.
Are you sure that it's not referring to Solaris 1.x? I don't think I've
ever seen a Solaris 2 box with any /dev/sbus. I don't have access to a
SunOS 4 box just at the moment to check that out, though.
Another possibility is that this support is offered by a third-party
driver.
ok
r.
8:30 a.m.
r. 'bear' stricklin wrote:
On Mon, 11 Dec 2000, Ram Meenakshisundaram wrote:
Are you sure that it's not referring to Solaris 1.x? I don't think I've
ever seen a Solaris 2 box with any /dev/sbus. I don't have access to a
SunOS 4 box just at the moment to check that out, though.
It is Solaris 2.6 (aka SunOS5) NOT Solaris 1.x (aka SunOS4). The BTSS
supports both SunOS4 and SunOS5. The
university machines that runs BTSS also runs it on Solaris 2.6, but I
cant seem to get my system up
and running while their's is. The card is the exact same card they have
and the same version of the OS.
The documentation for the BTSS supposedly has instructions on how to
create the devices, but I dont
have it nor do I have access to it either. Doing a "ls -l" on the
university machine's /dev shows the same
symbolic links I put it, but it doesnt work :-(
IPX
workstation running Solaris 2.6. I have the Basic Transputer
Support Software (BTSS 3.1) which has support for Solaris. However, I
do not have any documentation. The BTSS requires the creation
of "/dev/sbus?" in order for it to communicate with the transputer. I
do not have any /dev/sbus[0-3] devices on my system.
Another possibility is that this support is offered by a third-party
driver.
I find that hard to believe as there MUST be some way to get access to
the raw sbus via Solaris to
communicate with ANY sbus-based cards. And besides, the BTSS is the
device driver software, but
it needs the /dev/sbus[0-3] devices to communciate directly to the card....
Ram
9:09 a.m.
Another
possibility is that this support is offered by a third-party
driver.
I find that hard to believe as there MUST be some way to get access to
the raw sbus via Solaris to
communicate with ANY sbus-based cards. And besides, the BTSS is the
device driver software, but
it needs the /dev/sbus[0-3] devices to communciate directly to the card....
9:34 a.m.
On Mon, 11 Dec 2000, Ram Meenakshisundaram wrote:
It is Solaris 2.6 (aka SunOS5) NOT Solaris 1.x (aka
SunOS4). The BTSS
supports both SunOS4 and SunOS5. The university machines that runs
BTSS also runs it on Solaris 2.6, but I cant seem to get my system up
and running while their's is.
Ok, it was just a thought.
have it nor do I have access to it either. Doing a
"ls -l" on the
university machine's /dev shows the same symbolic links I put it, but
it doesnt work :-(
What type of machine is it running on at Uni? If it's not sun4c the path
to the pseudo-device in /devices may be different. It's possible even that
among sun4c machines the device paths may be different.
Contact me privately and I'll see if I can't help you out.
ok
r.
8:31 a.m.
"r. 'bear' stricklin" wrote:
[snip]
Are you sure that it's not referring to Solaris
1.x? I don't think I've
ever seen a Solaris 2 box with any /dev/sbus. I don't have access to a
SunOS 4 box just at the moment to check that out, though.
/dev/sbus isn't something one would expect to find under Solaris,
although you could fake it by creating a simlink to the appropriate
things in /devices. On an IPX I'd expect that to be something
like:
/devices/sbus@1,f8000000/sbusmem@x,0:slotx
Where <x> is the slot number, 0 - 3.
--
Chris Kennedy
chris(a)mainecoon.com
http://www.mainecoon.com
PGP fingerprint: 4E99 10B6 7253 B048 6685 6CBC 55E1 20A3 108D AB97
8:45 a.m.
Chris Kennedy wrote:
/dev/sbus isn't something one would expect to find
under Solaris,
although you could fake it by creating a simlink to the appropriate
things in /devices. On an IPX I'd expect that to be something
like:
/devices/sbus@1,f8000000/sbusmem@x,0:slotx
Where <x> is the slot number, 0 - 3.
That is exactly what I did, but it didnt work. Maybe I have some type
of hardware problem. I'll
have to check it out tonight. Thanks for all the help and responses.
Kudos to ClassicCmp....
Ram
9:02 a.m.
Ram Meenakshisundaram wrote:
That is exactly what I did, but it didnt work. Maybe
I have some type
of hardware problem. I'll
have to check it out tonight. Thanks for all the help and responses.
Kudos to ClassicCmp....
Um, if the device is showing up in /var/adm/messages after a reboot
then I'd take a hard look at /etc/path_to_inst. It's used to store
kernel metadata describing devices across boots...
--
Chris Kennedy
chris(a)mainecoon.com
http://www.mainecoon.com
PGP fingerprint: 4E99 10B6 7253 B048 6685 6CBC 55E1 20A3 108D AB97
8619
days inactive
8619
days old
15 comments
10 participants
participants (10)
-
chris＠mainecoon.com -
elvey＠hal.com -
korpela＠ellie.ssl.berkeley.edu -
mark＠cs.ualberta.ca -
red＠bears.org -
rich＠alcor.concordia.ca -
rivie＠teraglobal.com -
rmeenaks＠olf.com -
spectre＠stockholm.ptloma.edu -
THETechnoid＠home.com