From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Hellwig Subject: Re: [PATCH] vfs: add d_prune dentry operation Date: Mon, 10 Oct 2011 12:21:15 -0400 Message-ID: <20111010162115.GB751@infradead.org> References: <1317875170-9349-1-git-send-email-sage@newdream.net> <1317875170-9349-2-git-send-email-sage@newdream.net> <20111006212118.GA24109@infradead.org> <20111010051141.GV3159@dastard> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Dave Chinner , Christoph Hellwig , linux-fsdevel@vger.kernel.org, viro@ZenIV.linux.org.uk, ceph-devel@vger.kernel.org, rwheeler@redhat.com, linux-kernel@vger.kernel.org To: Sage Weil Return-path: Received: from 173-166-109-252-newengland.hfc.comcastbusiness.net ([173.166.109.252]:50707 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752703Ab1JJQVT (ORCPT ); Mon, 10 Oct 2011 12:21:19 -0400 Content-Disposition: inline In-Reply-To: Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Mon, Oct 10, 2011 at 09:19:43AM -0700, Sage Weil wrote: > Let me know if the below is clearer. The requirement is that ->d_prune be > called prior to unhashing (and then destroying) a victim dentry. I'm a > little worried about mixing that in with the lru helpers because it is > only indirectly related to whether the dentry is on the LRU, and that may > confuse people. A cleaned up opencoded patch is here: > > https://github.com/NewDreamNetwork/ceph-client/commit/784a6aa6dc7baf2069c40988d79130dba17c7068 > > and the updated dentry_lru_prune() wrapper version is below. Long term what I really want is a helper for doing more of the real removal in single place. This is one step towards it, and the series from Dave is another.