Return-Path: linux-nfs-owner@vger.kernel.org Received: from mail-bk0-f53.google.com ([209.85.214.53]:36422 "EHLO mail-bk0-f53.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751107Ab3HFOaU (ORCPT ); Tue, 6 Aug 2013 10:30:20 -0400 Received: by mail-bk0-f53.google.com with SMTP id d7so181411bkh.40 for ; Tue, 06 Aug 2013 07:30:18 -0700 (PDT) From: Miklos Szeredi To: rwheeler@redhat.com, avati@redhat.com, viro@ZenIV.linux.org.uk Cc: bfoster@redhat.com, dhowells@redhat.com, eparis@redhat.com, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, linux-nfs@vger.kernel.org, Trond.Myklebust@netapp.com, swhiteho@redhat.com, mszeredi@suse.cz Subject: [PATCH 0/4] [RFC] safely drop directory dentry on failed revalidate Date: Tue, 6 Aug 2013 16:29:59 +0200 Message-Id: <1375799403-28544-1-git-send-email-miklos@szeredi.hu> Sender: linux-nfs-owner@vger.kernel.org List-ID: Here's a series for fixing issues with d_drop on a directory dentry with children and adding support for such dropped directories in fuse. I tested the fuse part, but not the other filesystems (AFS, GFS2, NFS, SYSFS). The changes to filesystems are pretty trivial, though. Thanks, Miklos --- Anand Avati (1): fuse: drop dentry on failed revalidate Miklos Szeredi (3): vfs: check submounts and drop atomically vfs: check unlinked ancestors before mount fuse: use d_materialise_unique() --- fs/afs/dir.c | 6 +- fs/dcache.c | 159 +++++++++++++++++++++++++++++++++++++++++++++++++ fs/fuse/dir.c | 76 ++++++++++------------- fs/gfs2/dentry.c | 6 +- fs/internal.h | 1 + fs/namespace.c | 9 +++ fs/nfs/dir.c | 10 ++-- fs/sysfs/dir.c | 6 +- include/linux/dcache.h | 1 + 9 files changed, 216 insertions(+), 58 deletions(-)