Date: Mon, 3 Mar 1997 16:58:03 +0200 (IST)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
To: Christoph Kukulies <kuku AT gilberto DOT physik DOT RWTH-Aachen DOT DE>
cc: djgpp AT delorie DOT com
Subject: Re: dpmi_int - any caveats?
In-Reply-To: <199703031317.OAA11566@gil.physik.rwth-aachen.de>
Message-ID: <Pine.SUN.3.91.970303165321.9649C-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII


On Mon, 3 Mar 1997, Christoph Kukulies wrote:

> I read djgppfaq.txt chapter 18.2 and as a result I thought I
> had it working but the combo of my program and the driver
> seems to be very unstable. I cannot invoke the program twice.

Please post more of your code (the fragment you showed seems OK).  For 
example, how do you move data from and to the transfer buffer?

Also, can you see where does the program hang?  Is it in your code, or 
inside the call to `__dpmi_int'?

> BTW, why is __tb>>4 anded with 0xffff? Since it is an address below 1MB
> supplied by go32 the by 4 shifted value can never have any bits set
> that would be masked off by 0xffff.

You forget the memory mapping.  The transfer buffer is physically indeed
below 1MB mark, but its logical address is not guaranteed to have all the
high bits zeroed. 

> Does it matter whether cwsdpmi is loaded once in autoexec (cwsdpmi -p)
> or having it just in the path?

Shouldn't matter at all.