🚀 go-pugleaf

I have a TSR that will be invoked by the timer interrupt and at a certain
point in time, it will abort whatever the current process is and punch you
out to DOS. In other words, no matter what you are running, the TSR sits in
the background, and at a certain time kicks you out to DOS. (Why on earth 
would I want to do that? ... that's a whole different story)

Anyway, I'm using DOS interrupt 21h function 4Ch to terminate the current
process. It works fine for most cases. But when I'm running a graphics demo
program (FXSHOW or SOWPARTNER), it will punch me back to DOS ... but after
that I can not run the demo again. Most other programs work. COMMAND.COM
is still working (DOS commands work). But an editer I have also bombs out.

I think that the demo program is allocating memory that is not at its PSP
segment. So, those memory blocks are not getting deallocated.

Question is, is there a way to get a map, at any given time,  
of what memory blocks (segments) are allocated and who (what program) owns 
them?
Then maybe I can free those blocks allocated by my graphics program before
I murder it.

Any suggestions on this would be appreciated.

- Diptish Datta

From article <3108@tutor.tut.fi>, by th82787@tut.fi (Heinonen Tomi Tapani):
$ I have met a real problem with my 386 . It's not an IBM, it's clone.
$ However.. It has 1M base memory at morherboard (in SIMMs).
$ The system finds only 640 Kb. I have tried to change setups for move
$ memory that is betveen 640 Kb and 1M to location which is after 1M.
$ I haven't succeed. The dealer who sold that computer for me said that
$ it is impossible to use memory (betveen 640 K 1M) as an extendet or
$ expandet memory. I doubt that he is right.
$ If anyone knows something about that I would be gratefull if you
$ send me a mail and give some information :-)

Flame on:

this is c.s.i.p.PROGRAMMER. Your question belongs in comp.sys.ibm.pc.

Flame off:

Almost every 386 board I've seen is capable of mapping the 384K between
640K and 1MB above the 1MB mark. The AMI 386 board which I've seen in a
lot of clones only allows you to map 256K of the 384k.

What type of motherboard is it? 

j                               |%|John Lawitzke, Dale Computer Corp., R&D
                                |%|UUCP: uunet!frith!dale1!jhl         Work
                                |%|      uunet!frith!ipecac!jhl        Home
Inquiring minds just wondering. |%|Internet: jhl@frith.egr.msu.edu

In article <6570@cps3xx.UUCP> usenet@cps3xx.UUCP (Usenet file owner) writes:
>Almost every 386 board I've seen is capable of mapping the 384K between
>640K and 1MB above the 1MB mark. The AMI 386 board which I've seen in a
>lot of clones only allows you to map 256K of the 384k.

I have seen some cheap 286 clones which, when you have a Meg of ram installed,
will allow you to have 512k base + 512k extended or 640k base + 0 extended. 
Thus, if you use the second configuration, you waste 384k of perfectly good
memory, and there is no way you can get at it.  It may be that the 386 board
in question has the same stupid problem.  Unfortunatly, I don't remember what
brand or bios that machine was.

In article <8987@wpi.wpi.edu> ear@wpi.wpi.edu (Eric A Rasmussen) writes:
>
>I have seen some cheap 286 clones which, when you have a Meg of ram installed,
>will allow you to have 512k base + 512k extended or 640k base + 0 extended. 
>Thus, if you use the second configuration, you waste 384k of perfectly good
>memory, and there is no way you can get at it.  It may be that the 386 board

The Packard-Bell PB286 is one of these. Shame on P-B.  You either waste 384k of
memory, or limit yourself to 512k of base, and lock yourself out of many nice
programs.

Fernando

However with a 386 clone using this configuration you can get one of
the extended memory managers which will re-map the 384K wasted RAM
into useable ram. I've also seen a number of memory managers for the
286 that will do the same if you are using one of the later Phoenix
BIOS's.


                          ...Manu

I'm posting on behalf of Dr. M Strauss (MSTRAUSS @ CWRU)--
he is hard pressed to find a person to program cognitive psychology
experiments at Case Western Reserve University.  If you attend or
work at the university, or know anyone who fits the bill, please
respond to him.  Thanks.