From: Charles Sandmann <sandmann AT clio DOT rice DOT edu>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Weird mouse interrupt...Help!
Date: Wed, 27 Nov 2002 23:46:31 CST
Organization: Rice University, Houston TX
Lines: 36
Message-ID: <3de5adb7.sandmann@clio.rice.edu>
References: <hqs8uu04qrhp5363t9ref760hhbdstnn3i AT 4ax DOT com> <4d62e692 DOT 0211271946 DOT 799e233b AT posting DOT google DOT com>
NNTP-Posting-Host: clio.rice.edu
X-Trace: joe.rice.edu 1038462545 2182 128.42.105.3 (28 Nov 2002 05:49:05 GMT)
X-Complaints-To: abuse AT rice DOT edu
NNTP-Posting-Date: 28 Nov 2002 05:49:05 GMT
X-NewsEditor: ED-1.5.9
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

> When I replaced the source code disable() and enable() by 
> ("sti") and ("cli") in my H/W interrupt handler, then system 
> never been hung, they work fine both with PS/2 mouse and 
> COM port mouse. So I think my problem has been solved! but I don't know why?

Do *NOT* put an enable() call in your handler if you are using the
DJGPP wrappers.  Do *NOT* put a "sti" either.  The wrappers are
not re-entrant, and they restore interrupts on exit.

> -----Here is the snippet code captured from comp.os.msdos.djgpp-----
> #define INTERRUPT_ENTER(); asm("sti"); /* enable maskable interrupts*/
> #define INTERRUPT_LEAVE(); asm("cli"); /* disable maskable interrupts*/\
>                            outportb (0x20, 0x20); /* End of interrupt */
> #define INTERRUPT_LEAVE_BOTH();	asm ("cli"); /* disable maskable interrupts */ \
>                                 outportb (0x20, 0x20); /* End ofinterrupt  */ \
>                                 outportb (0xA0, 0x20); /* End ofinterrupt  */

> hardware_int_handler()
> {
>    INTERRUPT_ENTER();

>    //your handler code here.

>    INTERRUPT_LEAVE_BOTH();
> }

This is wrong; do not use the sti/cli