9 Mar
2005
9 Mar
'05
8:23 p.m.
So I imagine any device drive with the X at the end is
the better one to
use with the XM monitor? Or is it mandatory to use X device drivers in
the XM monitor?
It is mandatory... the system knows to append an X to any two-character
device name
So this brings up another question: I've noticed
that I don't explicitly
need to LOAD device drivers for the devices I want to use when I boot
RT-11. In particular, I don't need to LOAD DD when I want to use the
TU58 simulator. Are they found and loaded automatically when the OS
boots, or...?
Most utilities and user applications know to .FETCH device drivers they
need in order to do their task. After they are done, the drivers are
.RELEAsed. If they are LOADed before the program is executed, a .FETCH
and .RELEA are no-ops.
If they aren't loaded and aren't .FETCHed, then they simply are not in
memory.
Megan Gentry
Former RT-11 Developer
+--------------------------------+-------------------------------------+
| Megan Gentry, EMT/B, PP-ASEL,ST| email: mbg at world.std.com |
| Member of Technical Staff | megan at savaje.com |
| SavaJe Technologies, Inc. | (s/ at /@/) |
| 100 Apollo Drive | URL: http://world.std.com/~mbg/ |
| Chelmsford, MA 01824 | "pdp-11 programmer - some assembler |
| (978) 256 6521 (DEC '77-'98) | required." - mbg KB1FCA |
+--------------------------------+-------------------------------------+
12 Mar
12 Mar
6:36 p.m.
Megan wrote:
So I imagine
any device drive with the X at the end is the better one to
use with the XM monitor? Or is it mandatory to use X device drivers in
the XM monitor?
It is mandatory... the system knows to append an X to any two-character
device name
So this brings
up another question: I've noticed that I don't explicitly
need to LOAD device drivers for the devices I want to use when I boot
RT-11. In particular, I don't need to LOAD DD when I want to use the
TU58 simulator. Are they found and loaded automatically when the OS
boots, or...?
Most utilities and user applications know to .FETCH device drivers they
need in order to do their task. After they are done, the drivers are
.RELEAsed. If they are LOADed before the program is executed, a .FETCH
and .RELEA are no-ops.
If they aren't loaded and aren't .FETCHed, then
they simply are not in
memory.
Which does not mean the device drivers are unavailable,
simply not immediately ready. This means that with
device drivers such as DU(X).SYS using the RQDX1 or
RQDX2, it is best to use the LOAD command if repeated
use is expected since there are many seconds of delay
to initialize the controller. The RQDX3 is a bit faster.
NOTE that I agree with everything the Megan stated, I
was just adding a bit more detail.
Sincerely yours,
Jerome Fine
--
If you attempted to send a reply and the original e-mail
address has been discontinued due a high volume of junk
e-mail, then the semi-permanent e-mail address can be
obtained by replacing the four characters preceding the
'at' with the four digits of the current year.
13 Mar
13 Mar
9:16 a.m.
>It is mandatory... the system knows to append an X
to any two-character
>device name
For casual users of RT, this is the rule. For more experienced users,
there is a way of changing the 'X' to some other character. So, you
could have dd.SYS, for SB and FB, ddX.SYS for XB, XM, ddZ.SYS for ZB
and ZM, or some other mapping...
It is not trivial, as many drivers have the 'X' built in, but it is
possible.
Since Megan mentioned the monitors for V05.06, these
ddX.SYS
device drivers are also used with the RT11ZM. Note that both
RT11XM and RT11ZM are known as mapped monitors. In addition,
also starting with V05.06 (although I have never used them myself),
RT11XB and RT11ZB are also mapped, if I understand the
documentation correctly. I presume that these last two monitors
also use the ddX.SYS device drivers.
All the other monitors are unmapped and use the dd.SYS
device drivers.
Right...
One other minor point is that I seem to remember that
system jobs
are not allowed to perform .Fetch calls to the monitor. If a user
uses VBGEXE to run a normal program as a system job, then all
device handlers that are used must first be LOADed.
Good point... that is true. But it is possible for system and
foreground jobs to fetch handlers if certain steps are taken
and if the program is designed a certain way... but it is not
for the faint of heart. I believe I used the technique in my
RTEFTP program... I'll have to check on the specifics.
>If they aren't loaded and aren't .FETCHed,
then they simply are not in
>memory.
Which does not mean the device drivers are
unavailable,
simply not immediately ready. This means that with
Thanks... good point.
device drivers such as DU(X).SYS using the RQDX1 or
RQDX2, it is best to use the LOAD command if repeated
use is expected since there are many seconds of delay
to initialize the controller. The RQDX3 is a bit faster.
Megan Gentry
Former RT-11 Developer
+--------------------------------+-------------------------------------+
| Megan Gentry, EMT/B, PP-ASEL,ST| email: mbg at world.std.com |
| Member of Technical Staff | megan at savaje.com |
| SavaJe Technologies, Inc. | (s/ at /@/) |
| 100 Apollo Drive | URL: http://world.std.com/~mbg/ |
| Chelmsford, MA 01824 | "pdp-11 programmer - some assembler |
| (978) 256 6521 (DEC '77-'98) | required." - mbg KB1FCA |
+--------------------------------+-------------------------------------+
8:21 p.m.
Megan wrote:
Good point... that is true. But it is possible for
system and
foreground jobs to fetch handlers if certain steps are taken
and if the program is designed a certain way... but it is not
for the faint of heart. I believe I used the technique in my
RTEFTP program... I'll have to check on the specifics.
Jerome Fine replies:
I have never seen any documentation for this. I though
I remember seeing that the .Fetch EMT rejects a request
from a job that is not the background job (zero).
In addition, where is the device driver code to be placed,
especially if a background job is already running which
would often be the case?
Also, I asked about the possibility of DU(X).SYS having
128 or even 256 units. Did you miss that question?
Sincerely yours,
Jerome Fine
--
If you attempted to send a reply and the original e-mail
address has been discontinued due a high volume of junk
e-mail, then the semi-permanent e-mail address can be
obtained by replacing the four characters preceding the
'at' with the four digits of the current year.
7227
days inactive
7231
days old
3 comments
2 participants
participants (2)
-
jhfinexgs2＠compsys.to -
mbg＠TheWorld.com