From mboxrd@z Thu Jan 1 00:00:00 1970 From: Trond Myklebust Subject: Re: silent semantic changes with reiser4 Date: Sun, 29 Aug 2004 19:17:10 -0400 Sender: linux-fsdevel-owner@vger.kernel.org Message-ID: <1093821430.8099.49.camel@lade.trondhjem.org> References: <412F7D63.4000109@namesys.com> <20040827230857.69340aec.pj@sgi.com> <20040829150231.GE9471@alias> <4132205A.9080505@namesys.com> <20040829183629.GP21964@parcelfarce.linux.theplanet.co.uk> <20040829185744.GQ21964@parcelfarce.linux.theplanet.co.uk> <41323751.5000607@namesys.com> <20040829212700.GA16297@parcelfarce.linux.theplanet.co.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Alexander Viro , Hans Reiser , flx@msu.ru, Paul Jackson , riel@redhat.com, ninja@slaphack.com, diegocg@teleline.es, jamie@shareable.org, christophe@saout.de, vda@port.imtp.ilyichevsk.odessa.ua, christer@weinigel.se, spam@tnonline.net, Andrew Morton , wichert@wiggy.net, jra@samba.org, hch@lst.de, Linux Filesystem Development , linux-kernel@vger.kernel.org, flx@namesys.com, reiserfs-list@namesys.com Return-path: Received: from dh138.citi.umich.edu ([141.211.133.138]:27029 "EHLO lade.trondhjem.org") by vger.kernel.org with ESMTP id S268383AbUH2XSe convert rfc822-to-8bit (ORCPT ); Sun, 29 Aug 2004 19:18:34 -0400 To: Linus Torvalds In-Reply-To: List-Id: linux-fsdevel.vger.kernel.org P=E5 su , 29/08/2004 klokka 17:50, skreiv Linus Torvalds: > The general VFS layer has a lot of rules, and avoids these problems b= y > simply never having aliases between two directories. If the same dire= ctory > shows up multiple times (which can happen with bind mounts), they hav= e the > exact same dentry for the directory, it's just found through two diff= erent > vfsmount instances. That's why vfsmounts exist - they allow the same = name > cache entry to show up in different places at the same time. So could you explain what is stopping us from reducing the whole proble= m to the bind mount problem? IOW have "a/" be a directory that acts as if it is dynamically bind mounted on top of the file "a". Is it just the fantasy of supporting hard-links across "stream boundaries" (as in "touch a b; ln b a/b; ln a b/a")? I'm pretty sure nobody wants to have to add cyclic graph detection to their filesystems anyway. 8-) What other issues would need to be addressed? Cheers, Trond - To unsubscribe from this list: send the line "unsubscribe linux-fsdevel= " in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html