Date: Mon, 29 Aug 94 17:07:46 EDT
From: peprbv AT cfa0 DOT harvard DOT edu (Bob Babcock)
To: ELE AT CLIENTS DOT SWITCH DOT ch
Cc: djgpp AT sun DOT soe DOT clarkson DOT edu
Subject: Re: bugs in exec.c et al
Reply-To: babcock AT cfa DOT harvard DOT edu

> After using malloc with chunks > 4096 bytes, the free ram
> reported by _go32_dpmi_remaining_physical_memory() is not
> decreased accordingly.
> 
> Using calloc instead of malloc or writing to each page of the
> newly allocated space fixes that problem.
> 
> After returning from a call to spawnlp, in both cases the free
> ram reported does no longer reflect allocations done before.

When you do a spawn under dpmi, essentially all of your memory gets paged
out, and I would be surprised if it gets paged in again until it is
referenced.  So the above points are consistent, except that DJ says it isn't
supposed to work this way; _go32_dpmi_remaining_physical_memory() is supposed
to track allocations.