From:	"Alex Oleynikov" <alex AT compuweigh DOT com>
Newsgroups:	comp.os.msdos.djgpp
Subject:	Problem with tight kbhit() loop
Lines:	28
X-Priority:	3
X-MSMail-Priority:	Normal
X-Newsreader:	Microsoft Outlook Express 6.00.2600.0000
X-MimeOLE:	Produced By Microsoft MimeOLE V6.00.2600.0000
Message-ID:	<frN89.72711$Pb.3104967@news2.east.cox.net>
Date:	Wed, 21 Aug 2002 14:27:23 GMT
NNTP-Posting-Host:	68.15.54.9
X-Complaints-To:	abuse AT cox DOT net
X-Trace:	news2.east.cox.net 1029940043 68.15.54.9 (Wed, 21 Aug 2002 10:27:23 EDT)
NNTP-Posting-Date:	Wed, 21 Aug 2002 10:27:23 EDT
Organization:	Cox Communications
To:	djgpp AT delorie DOT com
DJ-Gateway:	from newsgroup comp.os.msdos.djgpp
Reply-To:	djgpp AT delorie DOT com

Hi all,

I have a question about using kbhit() function in a tight loop like the one
below:
...
while( !kbhit() );

ch = getch();
...

If I just run the program that contains this code everything works fine.
Now, if I run this program under the RHIDE's debugger and have some
breakpoints prior to the code listed above, then this loop never quits and
computer locks up completely (only cold re-boot helps to restart it) and
sometimes even my CMOS settings are get screwed up, so BIOS loads their
default values at start-up making me go nuts restoring them :o).

The problem goes away if I put some time-consuming operation inside the
loop, like a call to printf(..) or delay(..) functions.

Is there a rational explanation to this situation? Is it specific to DJGPP?
I have tried the same code under Borland Turbo C++ 2.0 and it always works
fine even if I have the breakpoints to visit before getting into that loop.

Thanks in advance,
Alex

- Raw text -