From: Con Kolivas <kernel@kolivas.org>
To: Timothy Miller <miller@techsource.com>
Cc: Charlie Baylis <cb-lkml@fish.zetnet.co.uk>, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] O12.2int for interactivity
Date: Wed, 6 Aug 2003 11:23:50 +1000 [thread overview]
Message-ID: <1060133030.3f3058a68e126@kolivas.org> (raw)
In-Reply-To: <3F303494.3030406@techsource.com>
Quoting Timothy Miller <miller@techsource.com>:
>
> The interactivity detection algorithm will always be inherently
> imperfect. Given that it is not psychic, it cannot tell in advance
> whether or not a given process is supposed to be interactive, so it must
> GUESS based on its behavior.
>
> Furthermore, for any given scheduler algorithm, it is ALWAYS possible to
> write a program which causes it to misbehave.
>
> This "thud" program is Goedel's theorem for the interactivity scheduler
> (well, that's not exactly right, but you get the idea). It breaks the
> system. If you redesign the scheduler to make "thud" work, then someone
> will write "thud2" (which is what you have just done!) which breaks the
> scheduler. Ad infinitum. It will never end. And in this case,
> optimizing for "thud" is likely also to make some other much more common
> situations WORSE.
>
>
> So, while it MAY be of value to write a few "thud" programs, don't let
> it go too far. The scheduler should optimize for REAL loads -- things
> that people actually DO. You will always be able to break it by
> reverse-engineering it and writing a program which violates its
> expectations. Don't worry about it. You will always be able to break
> it if you try hard enough.
Thank you for your commentary which I agree with. With respect to these
potential issues I have always worked on a fix for where I thought real world
applications might cause these rather than try and fix it for just that program.
It was actually the opposite reason that my patch prevented thud from working;
it is idle tasks that become suddenly cpu hogs that in the real world are
potential starvers, and I made a useful fix for that issue. Thud just happened
to simulate those conditions and I only tested for it after I heard of thud. So
just a (hopefully reassuring) reminder; I'm not making an xmms interactivity
estimator, nor an X estimator, nor a "fix this exploit" one and so on.
Con
next prev parent reply other threads:[~2003-08-06 1:23 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2003-08-04 19:50 [PATCH] O12.2int for interactivity Charlie Baylis
2003-08-05 2:10 ` Con Kolivas
2003-08-05 22:49 ` Timothy Miller
2003-08-06 0:12 ` charlie.baylis
2003-08-06 1:23 ` Con Kolivas [this message]
2003-08-06 22:24 ` Timothy Miller
2003-08-11 8:14 ` Rob Landley
2003-08-11 23:49 ` Timothy Miller
2003-08-12 0:17 ` William Lee Irwin III
2003-08-12 15:04 ` Timothy Miller
2003-08-12 23:32 ` William Lee Irwin III
2003-08-13 15:46 ` Timothy Miller
2003-08-14 6:09 ` William Lee Irwin III
2003-08-14 6:59 ` Con Kolivas
2003-08-14 7:01 ` William Lee Irwin III
2003-08-14 7:46 ` Con Kolivas
2003-08-14 20:03 ` Timothy Miller
2003-08-15 16:40 ` Con Kolivas
2003-08-14 20:00 ` Timothy Miller
2003-08-15 16:38 ` Con Kolivas
2003-08-15 18:12 ` Timothy Miller
2003-08-17 2:19 ` William Lee Irwin III
2003-08-17 18:00 ` Mike Fedyk
2003-08-14 19:57 ` Timothy Miller
2003-08-15 16:35 ` Con Kolivas
2003-08-15 18:17 ` Timothy Miller
2003-08-16 2:29 ` Con Kolivas
2003-08-14 19:54 ` Timothy Miller
-- strict thread matches above, loose matches on Subject: below --
2003-08-03 21:19 Voluspa
2003-08-04 2:34 ` Con Kolivas
2003-08-03 10:14 Con Kolivas
2003-08-03 11:25 ` Ingo Molnar
2003-08-03 11:36 ` Con Kolivas
2003-08-04 3:06 ` Con Kolivas
2003-08-03 11:37 ` Felipe Alfaro Solana
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=1060133030.3f3058a68e126@kolivas.org \
--to=kernel@kolivas.org \
--cc=cb-lkml@fish.zetnet.co.uk \
--cc=linux-kernel@vger.kernel.org \
--cc=miller@techsource.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;
as well as URLs for NNTP newsgroup(s).