From:	"Alex Vinokur" <alexvn AT bigfoot DOT com>
Newsgroups:	comp.os.msdos.djgpp
Subject:	Re: uclock() counts down?
Date:	Thu, 13 Feb 2003 18:46:37 +0200
Lines:	35
Message-ID:	<b2gi5i$1cmk6s$1@ID-79865.news.dfncis.de>
References:	<795DD3C2E97585449DCDEEE79CCD5C22BA8C AT email2k DOT compuweigh DOT com>
NNTP-Posting-Host:	pop09-2ras2-p120.barak.net.il (212.150.105.120)
X-Trace:	fu-berlin.de 1045154803 46878940 212.150.105.120 (16 [79865])
X-Priority:	3
X-MSMail-Priority:	Normal
X-Newsreader:	Microsoft Outlook Express 5.50.4522.1200
X-MimeOLE:	Produced By Microsoft MimeOLE V5.50.4522.1200
To:	djgpp AT delorie DOT com
DJ-Gateway:	from newsgroup comp.os.msdos.djgpp
Reply-To:	djgpp AT delorie DOT com

"Alex O" <Alex AT compuweigh DOT com> wrote in message news:795DD3C2E97585449DCDEEE79CCD5C22BA8C AT email2k DOT compuweigh DOT com...
> Hi everybody,
>
> I have noticed a strange behavior with uclock() in my program and was just wondering if there is a rational explanation to that.
> After some recent changes in the application I have noticed that the time-dependent values on my real-time graphs may flow
backwards! I use uclock() to time the change of certain parameters, because they are too fast for clock() (60-100 times per second),
and monitoring the return values of uclock() I saw that it in fact may sometimes return smaller values than before for a few calls
in a row!
> After about a day digging through the code I finally found this strange piece of code:
>
> uclock();
> ..
> inportb(0x40);
>
> which causes this odd behavior. If reading from the port 0x40 (system's timer register) is commented out or placed before the
uclock(), everything works OK.
>
> Any thoughts why?
>
> Alex O.
>

If you work with uclock() on Windows 2000 look at
http://groups.google.com/groups?th=f073ad9b3aca2a6a
http://groups.google.com/groups?th=ccb93720088dc36d

  ==============================
   Alex Vinokur
     mailto:alexvn AT go DOT to
     http://www.simtel.net/pub/oth/19088.html
   ==============================

- Raw text -