linux-kernel - Re: [RFC][PATCH 7/6] memcg: add atribute (for change bahavior ofrmdir)

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]

Date:	Thu, 6 Nov 2008 21:47:24 +0900 (JST)
From:	"KAMEZAWA Hiroyuki" <kamezawa.hiroyu@...fujitsu.com>
To:	"Hugh Dickins" <hugh@...itas.com>
Cc:	"KAMEZAWA Hiroyuki" <kamezawa.hiroyu@...fujitsu.com>,
	balbir@...ux.vnet.ibm.com,
	"linux-mm@...ck.org" <linux-mm@...ck.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"menage@...gle.com" <menage@...gle.com>,
	"nishimura@....nes.nec.co.jp" <nishimura@....nes.nec.co.jp>
Subject: Re: [RFC][PATCH 7/6] memcg: add atribute (for change bahavior 
     ofrmdir)

Hugh Dickins said:
> On Thu, 6 Nov 2008, KAMEZAWA Hiroyuki wrote:
>> On Thu, 06 Nov 2008 12:24:11 +0530
>> Balbir Singh <balbir@...ux.vnet.ibm.com> wrote:
>> > KAMEZAWA Hiroyuki wrote:
>> > >
>> > > 1. change force_empty to do move account rather than forget all
>> >
>> > I would like this to be selectable, please. We don't want to break
>> behaviour and
>> > not everyone would like to pay the cost of movement.
>>
>> How about a patch like this ? I'd like to move this as [2/7], if
>> possible.
>> It obviously needs painful rework. If I found it difficult, schedule
>> this as [7/7].
>>
>> BTW, cost of movement itself is not far from cost for force_empty.
>>
>> If you can't find why "forget" is bad, please consider one more day.
>
> My recollection from a year ago is that force_empty totally violated
> the rules established elsewhere, making a nonsense of it all: once a
> force_empty had been done, you couldn't really be sure of anything
> (perhaps it deserved a Taint flag).
>
yes. that was terrible. (but necessary to do so because we accounted
pages not on LRU or some other reason.)

> Without studying your proposals at all, I do believe you've a good
> chance of creating a sensible and consistent force_empty by moving
> account, and abandoning the old "forget all" approach completely.
>

yes. I'm very encouraged. thanks!
After patch [1/6]
  -> move all at force empty
After this
  -> move all or free (not forget) all.

-Kame


--
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/