[Coco] Memory leak in OS-9 Level Two

[KnudsenMJ at aol.com]

In a message dated 8/15/04 5:54:36 PM Eastern Daylight Time, 
boisy at boisypitre.com writes:

> At step 5, mfree should be 432K, but 8K seems to get lost.  In fact, if 
>  you follow steps 2-5 again, you will see mfree at step 5 go to 416K.  
>  You can do this every time, and 8K is lost each time.
>  
>  I don't recall anyone ever mentioning this particular problem.  Can 
>  someone verify that this is happening on their system?

I haven't booted up to check it (yet), but just from vague memory I have the 
following:
UltiMusE-III creates, inits, and uses a fresh window when it's run.  
(Originally that window was a dummy, just to base graphics blocks off of to use a 
virtual memory blocks, but later it was actually used for all graphics, after I 
kicked the VDG habit).

I could swear that after an evening of several runs of UME-III, the mfree 
seemed to have shrunk.  It may be that my window was leaking 8K each time.

BTW, there's a great utility, whose name I forget but I've used many times.  
It lists *every* 8K block in your system, who it's allocated to, and for what 
purpose.  I used it to check up on my UME virtual memory blocks.  Anyway, it 
would certainly pinpoint the "leaked" zombie windows.  Might give a clue as to 
why they're still allocated (or in fact they indicate as free, but somehow 
aren't).
--Mike K.