linux-kernel - Re: [RFC PATCH] Make swap accounting default behavior configurable

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 for Android: free password hash cracker in your pocket

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

Message-Id: <20101112094118.b02b669f.nishimura@mxp.nes.nec.co.jp>
Date:	Fri, 12 Nov 2010 09:41:18 +0900
From:	Daisuke Nishimura <nishimura@....nes.nec.co.jp>
To:	Michal Hocko <mhocko@...e.cz>
Cc:	linux-mm@...ck.org,
	KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>,
	linux-kernel@...r.kernel.org,
	Daisuke Nishimura <nishimura@....nes.nec.co.jp>
Subject: Re: [RFC PATCH] Make swap accounting default behavior configurable

On Thu, 11 Nov 2010 10:31:55 +0100
Michal Hocko <mhocko@...e.cz> wrote:

> On Thu 11-11-10 09:46:13, Daisuke Nishimura wrote:
> > On Wed, 10 Nov 2010 13:51:54 +0100
> > Michal Hocko <mhocko@...e.cz> wrote:
> > 
> > > Hi,
> > > could you consider the patch bellow? It basically changes the default
> > > swap accounting behavior (when it is turned on in configuration) to be
> > > configurable as well. 
> > > 
> > > The rationale is described in the patch but in short it makes it much
> > > more easier to enable this feature in distribution kernels as the
> > > functionality can be provided in the general purpose kernel (with the
> > > option disabled) without any drawbacks and interested users can enable
> > > it. This is not possible currently.
> > > 
> > > I am aware that boot command line parameter name change is not ideal but
> > > the original semantic wasn't good enough and I don't like
> > > noswapaccount=yes|no very much. 
> > > 
> > > If we really have to stick to it I can rework the patch to keep the name
> > > and just add the yes|no logic, though. Or we can keep the original one
> > > and add swapaccount paramete which would mean the oposite as the other
> > > one.
> > > 
> > hmm, I agree that current parameter name(noswapaccount) is not desirable
> > for yes|no, but IMHO changing the user interface(iow, making what worked before 
> > unusable) is worse.
> > 
> > Although I'm not sure how many people are using this parameter, I vote for
> > using "noswapaccount[=(yes|no)]".
> 
> Isn't a new swapaccount parameter better than that? I know we don't want
> to have too many parameters but having a something with a clear meaning
> is better IMO (noswapaccount=no doesn't sound very intuitive to me).
> 
Fair enough. It's just an trade-off between compatibility and understandability.

> > And you should update Documentation/kernel-parameters.txt too.
> 
> Yes, I am aware of that and will do that once there is an agreement on
> the patch itself. At this stage, I just wanted to have a feadback about
> the change.
> 
I'll ack your patch when it's been released with documentation update.

Thanks,
Daisuke Nishimura.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/