mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Antonio Vargas <wind@cocodriloo.com>
To: Hubertus Franke <frankeh@watson.ibm.com>
Cc: Antonio Vargas <wind@cocodriloo.com>,
	William Lee Irwin III <wli@holomorphy.com>,
	linux-kernel@vger.kernel.org, Robert Love <rml@tech9.net>,
	frankeh@optonline.net, nagar.us.ibm.com@elinux01.watson.ibm.com
Subject: Re: fairsched + O(1) process scheduler
Date: Tue, 8 Apr 2003 22:19:45 +0200	[thread overview]
Message-ID: <20030408201945.GB21496@wind.cocodriloo.com> (raw)
In-Reply-To: <200304081456.34367.frankeh@watson.ibm.com>

On Tue, Apr 08, 2003 at 02:56:34PM -0400, Hubertus Franke wrote:
> 
> Antonio, while you are coding here is some additional input.

Great!
 
> By intoducing the user based pending queue and leaking
> the tasks back into the runqueue based on the per user ticks, you
> are changing semantics slightly.
> 
> If the task is reinserted back into the runqueue oit should be 
> reinserted into the expired queue iff and only no
> expired/active queue switch has happened. 
> Otherwise it should be reinserted into the active list.

This sounds right :)
 
> This will becoe important when we later distinguish between
> users that have limited share and those that have unlimited.

Yes, we should push the "p->user->uid == 0" test into the
send_task_to_user() function then... and later on implement
a "p->user->unlimited_cpu_share == 1" test.
 
> Can be implemented by keeping a per runqueue array switch count
> and store it with the pending task. On reinsertion, if 
> the task has the same as the current it goes to the expired.
> If its older than the current, then the task missed the array switch
> and it should go into the active queue.
> 
> Also, I don't follow necessarily your reason to put an INTERACTIVE
> task back on the active queue immediately, rather than going first
> through the pending queue again.
> 
> This way, a high priority job with even a small sleep_avg already being 
> declared INTERACTIVE, will continue to suck up cycles beyond its
> user's limits. This can be as long as 8 secs currently.

Perhaps I'm not declaring this in any explicit way, but I feel
that this type of control should only be applied to cpu hogs.

> Instead, things to consider is to feed these as well through the 
> pending queue and distinguish in the pending queue ?
> More thoughts required here... I let you know when any of them is
> successful at my end.

I'm somewhat deadlocked right now, so I'll have to wait until I have
some think up some way to make it work as intended.
 
Greets, Antonio

  parent reply	other threads:[~2003-04-08 19:58 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-04-01 12:51 Antonio Vargas
2003-04-01 16:41 ` William Lee Irwin III
2003-04-01 22:19   ` Antonio Vargas
2003-04-02 12:46     ` Antonio Vargas
2003-04-02 16:35       ` William Lee Irwin III
2003-04-02 21:36         ` Antonio Vargas
2003-04-02 21:35           ` Robert Love
2003-04-02 22:07             ` Antonio Vargas
2003-04-02 22:10               ` Robert Love
2003-04-02 22:35                 ` Antonio Vargas
2003-04-03 17:15                 ` Corey Minyard
2003-04-03 18:17                   ` Robert Love
     [not found]     ` <200304021144.21924.frankeh@watson.ibm.com>
2003-04-03 12:53       ` Antonio Vargas
2003-04-03 19:22         ` William Lee Irwin III
2003-04-04 11:27           ` Antonio Vargas
2003-04-04 14:04             ` William Lee Irwin III
2003-04-04 20:12               ` Antonio Vargas
2003-04-04 20:40                 ` William Lee Irwin III
     [not found]             ` <200304041453.16630.frankeh@watson.ibm.com>
2003-04-04 21:36               ` Antonio Vargas
     [not found]                 ` <200304081456.34367.frankeh@watson.ibm.com>
2003-04-08 20:19                   ` Antonio Vargas [this message]
2003-04-01 18:33 ` Robert Love

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20030408201945.GB21496@wind.cocodriloo.com \
    --to=wind@cocodriloo.com \
    --cc=frankeh@optonline.net \
    --cc=frankeh@watson.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=nagar.us.ibm.com@elinux01.watson.ibm.com \
    --cc=rml@tech9.net \
    --cc=wli@holomorphy.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®