From: Dmitry Monakhov Subject: Re: [PATCH 08/11] ext4: introduce subtree logic Date: Tue, 09 Feb 2010 17:48:10 +0300 Message-ID: <87ljf2wjmd.fsf@openvz.org> References: <1265635693-12182-1-git-send-email-dmonakhov@openvz.org> <1265635693-12182-5-git-send-email-dmonakhov@openvz.org> <1265635693-12182-6-git-send-email-dmonakhov@openvz.org> <1265635693-12182-7-git-send-email-dmonakhov@openvz.org> <1265635693-12182-8-git-send-email-dmonakhov@openvz.org> <1265635693-12182-9-git-send-email-dmonakhov@openvz.org> <87f94c371002081554k3e4d6019q332602fea344ccc9@mail.gmail.com> <874olqpvtt.fsf@openvz.org> <87f94c371002090536v2450e67dlb77fa70bfc46cec0@mail.gmail.com> <87sk9awm66.fsf@openvz.org> <87f94c371002090629l6e496b23w8eeb7bbdf3e9b1fe@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: linux-ext4@vger.kernel.org, Jan Kara , OHSM-DEV To: Greg Freemyer Return-path: Received: from mail-bw0-f223.google.com ([209.85.218.223]:57494 "EHLO mail-bw0-f223.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754460Ab0BIOsP convert rfc822-to-8bit (ORCPT ); Tue, 9 Feb 2010 09:48:15 -0500 Received: by bwz23 with SMTP id 23so218467bwz.1 for ; Tue, 09 Feb 2010 06:48:13 -0800 (PST) In-Reply-To: <87f94c371002090629l6e496b23w8eeb7bbdf3e9b1fe@mail.gmail.com> (Greg Freemyer's message of "Tue, 9 Feb 2010 09:29:43 -0500") Sender: linux-ext4-owner@vger.kernel.org List-ID: Greg Freemyer writes: > On Tue, Feb 9, 2010 at 8:53 AM, Dmitry Monakhov wrote: >> Greg Freemyer writes: >> >>> On Tue, Feb 9, 2010 at 5:06 AM, Dmitry Monakhov wrote: >>>> Greg Freemyer writes: >>>> >>>>> On Mon, Feb 8, 2010 at 8:28 AM, Dmitry Monakhov wrote: >>>>>> * Abstract >>>>>> =C2=A0A subtree of a directory tree T is a tree consisting of a = directory >>>>>> =C2=A0(the subtree root) in T and all of its descendants in T. >>>>>> >>>>>> =C2=A0Subtree feature allows to create an isolated (from user po= int of view) >>>>>> =C2=A0trees. >>>>>> >>>>>> =C2=A0Subtree assumptions: >>>>>> =C2=A0(1) Each inode has subtree id. This id is persistently sto= red inside >>>>>> =C2=A0 =C2=A0 =C2=A0inode (xattr, usually inside ibody) >>>>>> =C2=A0(2) Subtree id is inherent from parent directory >>>>>> =C2=A0(3) Inode can not belongs to different subtree >>>>>> =C2=A0 =C2=A0 =C2=A0Otherwise changes in one subtree result in c= hanges in other subtree >>>>>> =C2=A0 =C2=A0 =C2=A0which contradict to isolation criteria. >>>>>> >>>>>> =C2=A0This feature is similar to project-id in XFS. One may assi= gn some id to >>>>>> =C2=A0a subtree. Each entry from the subtree may be accounted in= directory >>>>>> =C2=A0subtree quota. Will appear in later patches. >>>>>> >>>>>> * Disk layout >>>>>> =C2=A0Subtree id is stored on disk inside xattr usually inside i= body. >>>>>> =C2=A0Xattr is used only as a data storage, It has not user visi= able xattr >>>>>> =C2=A0interface. >>>>>> >>>>>> Signed-off-by: Dmitry Monakhov >>>>> >>>>> Dmitry, >>>>> >>>>> I think the idea of subtrees is useful, but I'm curious about oth= er >>>>> use cases than just quota. >>>>> >>>>> At first glance you are attempting to create a generic subtree >>>>> functionality for ext4, but criteria 3) above says a inode can on= ly be >>>>> in one subtree at a time. >>>> Theoretically this is possible, but this dramatically complicate t= hings >>>> Just think about this. If inode belongs to different subtrees then >>>> it must have several tree-dquota objects attached to it. This mean= s >>>> that quota require great quota redesign. >>>> Obviously i don't know any use case for this feature. do you know = any? >>> >>> Maybe we're talking about different things. =C2=A0I working with th= e OHSM >>> project >>> >>> We haven't submitted any patches yet, but for one of our features w= e >>> have something fairly close to your subtree patch. =C2=A0If we were= to >>> leverage your patch and drop that part of ours we would be in the >>> unhappy situation that quota and ohsm could not both be enabled on = the >>> same filesystem because the ohsm subtree geography is not likely to= be >>> consistent with the quota subtree geography. >>> >>> If we call quota and ohsm services then my desire would be to see y= our >>> subtree patches support orthoganal subtree groups. =C2=A0One group = per >>> service. >>> >>> I haven't looked into the actual implementation, but from an API >>> perspective it is just a matter of adding a service parameter to th= e >>> various calls. =C2=A0For a given subtree service group, a given ino= de could >>> only be part of one subtree, but a single inode could participate i= n >>> multiple subtree service groups. >> Ok now i think i understand what you are talking about. >> one subtree <=3D> one service is the main rule i'm standing. >> If you wan to support several services, no problem >> I can easily extend xattr to support different services >> something like this >> subtree_entry >> { >> =C2=A0 =C2=A0 =C2=A0 =C2=A0__le16 sbe_flags /* entry flags */ >> =C2=A0 =C2=A0 =C2=A0 =C2=A0__le16 sbe_type =C2=A0/* service type */ >> =C2=A0 =C2=A0 =C2=A0 =C2=A0__le32 sbe_id =C2=A0 =C2=A0/* subtree id = */ >> } > > Conceptually agreed. > > But you'd still be limited to one subtree_entry per inode with that r= ight? > > Another option would be more like > > #define EXT4_SUBTREE_MAX_SERVICE_GROUPS 1 > #define EXT4_SUBTREE_MAX_SERVICE_GROUP_QUOTA 0 > > // OHSM would patch the above when its ready to submit > > subtree_entry[EXT4_SUBTREE_MAX_SERVICE_GROUPS] > { > __le16 sbe_flags /* entry flags */ > __le32 sbe_id /* subtree id */ > } > > (Clearly that won't actually work as is, but you get the idea. Each > service group gets its own flags and id.) > > Another option would be separate xattr entries for each subtree group= =2E > > Regardless, it looks very doable and its just a matter of figuring ou= t > the best way to support multiple subtree service groups. Off course i mean that. In fact i'm plan to pack entries in to one xattr data block because each xattr has significant space overhead. -- To unsubscribe from this list: send the line "unsubscribe linux-ext4" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html