Re: [PATCH] O16int for interactivity

From: Timothy Miller <miller@techsource.com>
To: Nick Piggin <piggin@cyberone.com.au>
Cc: Con Kolivas <kernel@kolivas.org>,
	linux kernel mailing list <linux-kernel@vger.kernel.org>,
	Andrew Morton <akpm@osdl.org>, Ingo Molnar <mingo@elte.hu>,
	gaxt <gaxt@rogers.com>, Mike Galbraith <efault@gmx.de>
Subject: Re: [PATCH] O16int for interactivity
Date: Mon, 18 Aug 2003 15:48:46 -0400	[thread overview]
Message-ID: <3F412D9E.4070807@techsource.com> (raw)
In-Reply-To: 3F40F43A.4050709@cyberone.com.au

Nick Piggin wrote:

>>   If another task exists at a higher priority, then it gets run at 
>> that point.
> 
> 
> 
> Well loosely, yes. Actually, it happens if the task exists and is 
> "running",
> and has timeslice left. 

> That only happens in scheduler_tick when the 
> current
> task has finished its timeslice and the priority arrays are about to be
> switched. 

What only happens then?

I'm confused again.  Are you talking about swapping the active and 
expired arrays?

Of course, all bets are off if the current task actually uses up it 
whole timeslice.  Then it's not being preempted in quite the same way.

So, then, if there are not tasks left in the active array, naturally, 
the highest priority task from what was once the expired array will be 
run, and that may be of higher priority.

Is that what you're saying?

 > The required conditions for preemption can also occur when a task
> is being woken up, (after sleeping or newly forked).

This is the case that I was thinking of.  No swapping of queues.  It's 
just that a higher priority task was sleeping (or not existing) which 
could cause the current task to be preempted before its timeslice ends.