From: Martin Str|mberg Message-Id: <200004250447.GAA02932@father.ludd.luth.se> Subject: Re: The new cwsdpmi In-Reply-To: <10004250411.AA12206@clio.rice.edu> from Charles Sandmann at "Apr 24, 2000 11:11:55 pm" To: sandmann AT clio DOT rice DOT edu (Charles Sandmann) Date: Tue, 25 Apr 2000 06:47:40 +0200 (MET DST) Cc: djgpp-workers AT delorie DOT com (DJGPP-WORKERS) X-Mailer: ELM [version 2.4ME+ PL54 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Reply-To: djgpp-workers AT delorie DOT com Errors-To: nobody AT delorie DOT com X-Mailing-List: djgpp-workers AT delorie DOT com X-Unsubscribes-To: listserv AT delorie DOT com Precedence: bulk Please CC djgpp-workers, like I did, as it seems to be good information for others too. According to Charles Sandmann: > I suggest you go back to the original CWSDPMI for a while and establish > everything works as you expect in a stable manner. Yepp. I'm working on it, although as it's not the fastest box on earth it takes a while... > Then just rename the CWSDPMI.EXE and replace it with the newer one. > > Does this system use emm386 or equivalent (VCPI) or is it limited to > just HIMEM.SYS? Just HIMEM.SYS. CONFIG.SYS: DEVICE=C:\DOS\HIMEM.SYS /v rem DEVICE=C:\DOS\SETVER.EXE DOS=HIGH COUNTRY=046,850,C:\DOS\COUNTRY.SYS DEVICE=C:\DOS\DISPLAY.SYS CON=(EGA,,1) FILES=30 STACKS=9,256 device=C:\NET\ifshlp.sys LASTDRIVE=Z AUTOEXEC.BAT: C:\NET\net initialize C:\DOS\SMARTDRV.EXE c+ d+ rem @ECHO OFF PROMPT $p$g PATH C:\WINDOWS;C:\NET;C:\DOS SET TEMP=C:\tmp MODE CON CODEPAGE PREPARE=((850) C:\DOS\EGA.CPI) MODE CON CODEPAGE SELECT=850 KEYB SV,,C:\DOS\KEYBOARD.SYS doskey /insert call netstart.bat call netuse.bat set PATH=e:\compress;e:\nc;%PATH% SET TEMP=c:\tmp SET TMP=c:\tmp set TMPDIR=c:/tmp/ set DJGPP=d:/djgpp/djgpp.env set PATH=d:\djgpp\bin;d:\djgpp\gnu\emacs\bin;%PATH% set LFN=y set SHELL=d:/djgpp/bin/bash.exe rem cwsdpmi.exe -p -st:\cwsdpmi.swp > > Once I had an error message saying "FPU overrun", I think. > > Very strange - this means the FPU had an exception (only found on 386 > chips). I've never seen one. Note that this is a 386 without any math coprocessor. > > But finally something very strange happened and I had still some > > prompt functionality and it seems (to me) to be something strange with > > CWSDPMI. Error message was "Swap disk full!", > > It's very possible there is some internal corruption in CWSDPMI which is > trashing the page bitmap or something - which means there's some ugly bug. > Maybe only seen on small memory systems. I don't know. > > > Any ideas? Do you want me to try something specific? > > Can you make a CWSDPMI with debug printouts available? > > It's tough to isolate unless it's reproducible. I probably need to send Well it's easy to reproduce strange events, but exact the reported behaviour might be more difficult. > my diagnostic routine. CWSDPMI with diagnostics is ugly - there is a Put it one the ftp site! > very simple printout I often display which doesn't tell much, plus > verbose stuff which if sent to the screen makes anything run with it > useless - or if it goes to a file you must disable hardware interrupts > or risk trashing your hard drive. That's ok. The most valuable things on this box's hard disk is probably the net setup... > No easy fix here. What's worse - I just noticed the binary I distributed > is not the most recent one on my hard drive :-( Let us know when you've updated it. Perhaps an announcement on c.o.m.d would be good too (for other people not reading djgpp-workers). Right, MartinS