From: Waldemar Schultz <schultz AT ma DOT tum DOT de>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: CWSDPMI crash
Date: Tue, 28 Sep 1999 18:33:33 +0200
Organization: [posted via] Leibniz-Rechenzentrum, Muenchen (Germany)
Lines: 75
Distribution: world
Message-ID: <37F0EDDD.F35EEA21@ma.tum.de>
References: <37EF9A4F DOT FEC8240 AT ma DOT tum DOT de> <37f003e2 DOT sandmann AT clio DOT rice DOT edu>
NNTP-Posting-Host: pcritter1.mathematik.tu-muenchen.de
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-Mailer: Mozilla 4.6 [de] (Win95; I)
X-Accept-Language: de,en-US
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

Charles Sandmann schrieb:

> >  31 MB  200e84c
>
> (crash)
>
> > Paragraphs of memory for extra CWSDPMI internal heap ? [16384]
>
> Too much.. should be room for static data
>

I set it to 8192 now. O.K.?


>
> > Maximum number of 4K pages in swap file ? [32767]
>
> Decrease this ... start with zero as a test (disable paging).
>

I decreased it so that total virtual memory becomes
less than 128 KB (seemed to be necessary)

>
> > DPMI memory available: 33285 Kb
> > DPMI swap space available: 129911 Kb
>
> Why is there only 32MB showing available?  very bad if ..  ah, don't use
> EMM386 unless you put NOVCPI on the command line...  There are many bad
> bugs in EMM386 if you reach it's max limit!  I see hard hangs on the
> W9x release at the limit.  Maybe fix in r5.
>

unfortunately I have to use EMM386 with VCPI because my project(makefile)
involes three different compilers that need it  :-(

>
> > mem.exe (BTW the machine runs a PII with 128 MB Ram installed !! ??)
>
> > ==========================================================
> > MS-DOS Version 6.22
>
> You should get the HIMEM.SYS from Win9x - it will allow you to use all
> of your memory.

I already use HIMEM.SYS fom W95 (24/Aug/1995)
no idea where the wiered reports by mem.exe come from  :-(


Now it seems to work anyway !!!
you can surely imagine that ALLOC.C wasn't really the program I had to fix
:-)

so many many thanks for help and of course for the smart job
you did for us supplying CWSDPMI

--


 Gruss Waldemar Schultz.


 ________________________________________

 Waldemar Schultz
 Technische Universität München
 Zentrum Mathematik M1, SCA
 D 80290 München

 Tel: +49 (0)89 2892 8226
 FAX: +49 (0)89 2892 8228
 e-mail: schultz AT ma DOT tum DOT de
 ________________________________________