From:	sparhawk AT eunet DOT at (Gerhard Gruber)
Newsgroups:	comp.os.msdos.djgpp
Subject:	Re: bash eats cycles
Date:	Sun, 05 Jul 1998 18:01:17 GMT
Organization:	Customer of EUnet Austria
Message-ID:	<35a7b81a.4043791@news.Austria.EU.net>
References:	<Pine DOT SUN DOT 3 DOT 91 DOT 980702163208 DOT 17900B-100000 AT is>
NNTP-Posting-Host:	e198.dynamic.vienna.at.eu.net
Mime-Version:	1.0
Lines:	39
To:	djgpp AT delorie DOT com
DJ-Gateway:	from newsgroup comp.os.msdos.djgpp

Destination: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
From: Gruber Gerhard
Group: comp.os.msdos.djgpp
Date: Thu, 2 Jul 1998 16:37:36 +0300 (IDT):

>On Thu, 2 Jul 1998, Gerhard W. Gruber wrote:
>
>> It's NTVDM.EXE that takes all the time.

I think that is the subssystem that manages 80x86 virtual machines (but I'm
not sure on this). At least all 16bit applications run under this.

>What's this?  (I'm not an NT guru, so go slow on me ;-).

Me neither. :) The longer I'm forced to work with it the less I like it. :)

>No, AFAIK Bash doesn't know when you minimize its window.  It seems like
>NT itself doesn't give Bash any cycles because it NT knows it's minimized,
>that's why minimizing the window remedies the situation. 

That's possible. I'm not sure if a DOS programm could even be aware of
max-/minimizing a DOS window.

>No, you don't.  You can write a simple program that waits for keys in a
>loop that calls __dpmi_yield when kbhit returns zero.  You can link this
>program once with the version of __dpmi_yield in the library and once 
>with the version I sent, and compare the CPU load imposed by both 
>programs.

That's a good idea. I'll try it out tomorrow.

--
Bye,
   Gerhard

email: sparhawk AT eunet DOT at
       g DOT gruber AT sis DOT co DOT at

Spelling corrections are appreciated.

- Raw text -