Date:	Sun, 9 Apr 2000 08:47:25 +0200 (IST)
From:	Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender:	eliz AT is
To:	Willem Bekker <bekker AT mweb DOT com DOT na>
cc:	djgpp AT delorie DOT com
Subject:	Re: HARDWARE INTERRUPT HANDLING BY CWSDPMI
In-Reply-To:	<E12cRZ6-0008Dx-00@mail.mweb.com.na>
Message-ID:	<Pine.SUN.3.91.1000409084705.7469T-100000@is>
MIME-Version:	1.0
Reply-To:	djgpp AT delorie DOT com
Errors-To:	nobody AT delorie DOT com
X-Mailing-List:	djgpp AT delorie DOT com
X-Unsubscribes-To:	listserv AT delorie DOT com

On Tue, 4 Apr 2000, Willem Bekker wrote:

> I have a small question.  How much does the handling of real mode interrupt
> handlers slow down when a djgpp program using cwsdpmi is running.  If, for
> example, I have a standard tsr handling com port interrupts at a high rate
> with real mode code, would a program like lynx under dos cause the system to
> loose interrupts, while a real mod program would not.

A DJGPP program indeed incurs additional overhead, because under DPMI,
all hardware interrupts are reflected to protected-mode handlers
first, and only if unhandled, they are passed to real-mode handlers.
The mode switch that this involes takes up hundreds of CPU cycles.

Whether this overhead will cause you to lose interrupts, I don't know.
In general, I won't expect the overhead to be severe enough, so the
system might as well cope with the additional load.

> Another question is how stable would a system be with a tsr requiring
> cwsdpmi.  For example would a system with such a tsr survive the starting up
> and the shutting down of Windows.

You cannot start Windows when a DJGPP program is running, because the
CPU is already in protected mode.  Windows will refuse to start.

- Raw text -