Message-ID: <3427709F.1212@EnchantedLearning.com>
Date: Tue, 23 Sep 1997 00:33:34 -0700
From: Mitchell Spector <spector AT EnchantedLearning DOT com>
Reply-To: spector AT EnchantedLearning DOT com
Organization: Enchanted Learning Software
MIME-Version: 1.0
To: djgpp AT delorie DOT com,
        "Salvador Eduardo Tropea (SET)" <salvador AT natacha DOT inti DOT edu DOT 2alpha DOT com>,
        Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
Subject: Clipboard timing, strange results, Rhide (Was:  DJGPP, interprocess communication, and DPMI)
References: <m0xD6Or-000S1kC AT inti DOT edu DOT ar>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Precedence: bulk

I tried timing the WINOLDAP clipboard interrupts, and the results
surprised me.  (Timing routines often do that. :-)

   The loop below will repeatedly open and close the Windows
clipboard for 10 seconds, counting the number of iterations.
Any useful work with the clipboard (getting the data or its
size, or setting the data) is over and above this, but this
is at least a baseline.

   Save the code below as clip1.cc.  I compiled it with the command
" gcc -Wall -o clip1.exe clip1.cc ".  The machine I tested it on
has a 133-MHz Pentium processor -- just a reference point for anybody
else who tries it.

   I ran it in a DOS box under Windows95, and found that I got
40-50 iterations per second.  That's slow, but that's not the
surprising part.  The surprising part came when I launched
Rhide, spawned a DOS shell from within Rhide (using the DOS shell
menu item), and found that I got between 700 and 850 iterations per
second!

   This is completely repeatable.  Running the program by
double-clicking on the clip1.exe icon yields about 50 iterations
per second; also starting a DOS box from the Start menu and running
clip1 from the command line in that DOS box yields about 50
iterations per second.  But starting Rhide in Win95, spawning a
DOS box from Rhide, and running clip1 from the command line in
that DOS box consistently yields around 800 iterations per second.

   Does anybody know what difference there is between a DOS box
spawned directly by Win95 and a DOS shell spawned by Rhide that
could account for a 15-fold speed improvement when running
under Rhide's DOS shell?  Are interrupts handled differently
in some way?

   Notice that I don't have to compile using Rhide, so it can't
be my current Rhide optimization settings, for instance.  I'm just
using Rhide as a gateway to a DOS shell and then running literally
the same exe that I ran before, only with much faster execution of
the interrupts.

   I checked the various PIF's, but I don't see any differences
between them (the PIF memory settings for DPMI memory, etc.,
are all Win95's stock "Auto" setting).

   When I remove the interrupts from the program entirely,
leaving just the overhead (loop, timing, and counting), the
difference disappears, and I get around 3 million iterations
per second, whether via Rhide or not.  (Incidentally, this shows
that the overhead is insignificant and that the interrupts
themselves are very, very slow.)

   Any ideas?  (See the code at the end of this posting, after
my signature.)

Mitchell
--
Mitchell Spector, Enchanted Learning Software
E-mail: spector AT EnchantedLearning DOT com
Visit ZoomDinosaurs, the first hypertext book for children about
dinosaurs.  -- http://www.EnchantedLearning.com/subjects/dinosaurs --


Here's the code:


// clip1.cc
#include <dpmi.h>
#include <stdio.h>
#include <time.h>

const int IntWin = 0x2f;
const int InterruptFunctionIdentifyWinOldApVersion = 0x1700;
const int InterruptFunctionOpenClipboard           = 0x1701;
const int InterruptFunctionEmptyClipboard          = 0x1702;
const int InterruptFunctionSetClipboardData        = 0x1703;
const int InterruptFunctionGetClipboardDataSize    = 0x1704;
const int InterruptFunctionGetClipboardData        = 0x1705;
const int InterruptFunctionCloseClipboard          = 0x1708;

const int ClipboardFormatText = 1;
const int WinOldApBadVersion = 0x1700;

const int NumberOfSeconds = 10; // Duration of test

int main(void)
    {
     __dpmi_regs r;
     clock_t ClockGoal;
     const clock_t durationInTicks = CLOCKS_PER_SEC * NumberOfSeconds; 
     long count = 0;
    
     r.x.ax = InterruptFunctionIdentifyWinOldApVersion;
     __dpmi_int(IntWin, &r);
     
     if (WinOldApBadVersion == r.x.ax)
        {
         printf("Doesn't support WINOLDAP clipboard interrupts\n");
         return 1;
        }
     
     printf("WINOLDAP Version %hu.%hu\n", r.h.al, r.h.ah);
 
	 ClockGoal = clock() + durationInTicks;
     
     while (clock() < ClockGoal)
        {
	     r.x.ax = InterruptFunctionOpenClipboard;
	     __dpmi_int(IntWin, &r);
	     
	     if (0 == r.x.ax)
	        {
	         printf("Error: Opening clipboard\n");
             return 2;
            }
         

	     r.x.ax = InterruptFunctionCloseClipboard;
	     __dpmi_int(IntWin, &r);
	     
	     if (0 == r.x.ax)
	        {
	         printf("Error: Closing clipboard\n");
             return 3;
            }
         
         count++;
        }          	  
     
	 printf("Opened and closed clipboard approximately %ld times per
sec\n", (count + NumberOfSeconds/2)/NumberOfSeconds);

     return 0;
    } // main