From: Ville Muikkula <vmuikkul AT ratol DOT fi>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Is PMODE/DJ Free Software?
Date: Sat, 29 Jun 2002 17:08:17 +0000 (UTC)
Organization: Oulu Polytechnic, Computing Centre
Lines: 28
Message-ID: <afkpi1$n58$1@pan.oamk.fi>
References: <afctgd$723$1 AT pan DOT oamk DOT fi> <3d1a87f4 DOT sandmann AT clio DOT rice DOT edu> <affdef$6i9$1 AT pan DOT oamk DOT fi> <3d1d5667 DOT sandmann AT clio DOT rice DOT edu>
NNTP-Posting-Host: ruutana.ratol.fi
X-Trace: pan.oamk.fi 1025370497 23720 193.167.144.11 (29 Jun 2002 17:08:17 GMT)
X-Complaints-To: usenet AT pan DOT oamk DOT fi
NNTP-Posting-Date: Sat, 29 Jun 2002 17:08:17 +0000 (UTC)
User-Agent: tin/1.4.5-20010409 ("One More Nightmare") (UNIX) (SunOS/5.7 (sun4u))
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

Charles Sandmann <sandmann AT clio DOT rice DOT edu> wrote:

> "It may be used or distributed in any manner you wish, as long as you do 
> not try to sell an extender based on PMODE."

  I think the intention of this is to prohibit anyone from making 
commercial, closed source derivatives, but it also takes away the right to 
sell (modified) copies, which is part of the definition of free software 
(by FSF). There are free software licenses that accomplish the first 
effect without the second one. Of course, he who writes the code gets to 
choose the license.

> PMODE/DJ has some mode switches which are about 2X faster in hardware
> interrupt handling than CWSDPR0.

  Is there any difference in IRQ handling performance between CWSDPR0 
and PMODE/DJ when one uses only protected mode interrupt service routines?

> If you have a high interrupt rate (> 10,000 interrupts per second) you 
> should definitely play with PMODE/DJ.

  1541EMU uses the LPT port IRQ for detecting changes on the Commodore 
serial bus /ATN signal state. Some Commodore 64 fastloaders use this 
signal as a bit clock and currently the interrupt handling overhead 
causes the emulation to occasionally fall behind enough to make it 
incompatible with such fastloaders. It seems to me that this can only be 
solved by using the Automatic EOI mode of PIC, because the timing is 
critical all the way down to single digit microsecond range.