linux-kernel - Re: RSDL v0.31

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-id: <200703171444.51314.jos@mijnkamer.nl>
Date:	Sat, 17 Mar 2007 14:44:48 +0100
From:	jos poortvliet <jos@...nkamer.nl>
To:	Ingo Molnar <mingo@...e.hu>
Cc:	Mike Galbraith <efault@....de>, Con Kolivas <kernel@...ivas.org>,
	linux-kernel@...r.kernel.org,
	Andrew Morton <akpm@...ux-foundation.org>,
	Linus Torvalds <torvalds@...ux-foundation.org>
Subject: Re: RSDL v0.31

Op Saturday 17 March 2007, schreef Ingo Molnar:
> * jos poortvliet <jos@...nkamer.nl> wrote:
> > Op Saturday 17 March 2007, schreef Ingo Molnar:
> > > so it is not at all clear to me that RSDL is indeed an improvement,
> > > if it does not have comparable auto-nice properties.
> >
> > Wasn't the point of RSDL to get rid of the auto-nice, because it
> > caused starvation, unpredictable behaviour and other problems?
>
> it doesnt really get rid of it, it replaces it with another mechanism
> that is fundamentally unfair too.
>
> RSDL has _another_, albeit more hidden "auto-nice" behavior: this time
> expressed not via the plain manipulation of priorities based on the
> sleep average, but expressed via the quota-depletion flux of tasks over
> time, fed into a complex dance of rotating priorities - which
> quota-depletion flux is in essence a sleep average too, just more
> derived and more hardcoded.

Hmmm. I wonder, then, does RSDL give an advantage in the areas I mentioned 
(starvation and predictability)? 

RSDL does give equal timeslices (eg equal cpu time) to each process - it's 
just that processes which didn't use their time yet can quickly run, right? 
Now I might not understand things here, but that it sounds more fair, though 
you're more quallified to judge that. So, for me, as an user, it boils down 
to: does this solve problems, and does it introduce problems?

I must say I compare RSDL to staircase, as that's what I'm used to - a bit 
more interactive compared to mainline. RSDL does slightly worse AND slightly 
better - worse in interactivity on heavy loads (8 makes running on my 
dualcore) but it doesn't have the systemwide stalls sometimes occurring on 
both mainline and staircase - though it replaces them with shorter, 
app-specific ones (the worse interactivity I mention).

> or looking at it from another angle, code size:
>
>    text    data     bss     dec     hex filename
>   15750      24    6008   21782    5516 sched.o.vanilla
>   15960     360    6336   22656    5880 sched.o.rsdl
>
> there's no reduction in complexity, it just moved elsewhere.
>
> 	Ingo

-- 
Disclaimer:

Alles wat ik doe denk en zeg is gebaseerd op het wereldbeeld wat ik nu heb. 
Ik ben niet verantwoordelijk voor wijzigingen van de wereld, of het beeld wat 
ik daarvan heb, noch voor de daaruit voortvloeiende gedragingen van mezelf. 
Alles wat ik zeg is aardig bedoeld, tenzij expliciet vermeld.

Content of type "application/pgp-signature" skipped