From mboxrd@z Thu Jan 1 00:00:00 1970 From: Al Viro Subject: Re: [PATCH 02/11] 9p: fix dentry leak in v9fs_vfs_atomic_open_dotl() Date: Tue, 17 Sep 2013 00:28:10 +0100 Message-ID: <20130916232810.GG13318@ZenIV.linux.org.uk> References: <1379335925-30858-1-git-send-email-miklos@szeredi.hu> <1379335925-30858-3-git-send-email-miklos@szeredi.hu> <20130916181938.GD13318@ZenIV.linux.org.uk> <20130916195003.GE13318@ZenIV.linux.org.uk> <20130916220241.GF13318@ZenIV.linux.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Linux-Fsdevel , Kernel Mailing List , "mszeredi@suse.cz" , Eric Van Hensbergen , "M. Mohan Kumar" , stable@vger.kernel.org To: Miklos Szeredi Return-path: Content-Disposition: inline In-Reply-To: <20130916220241.GF13318@ZenIV.linux.org.uk> Sender: stable-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On Mon, Sep 16, 2013 at 11:02:41PM +0100, Al Viro wrote: > On Mon, Sep 16, 2013 at 10:09:46PM +0200, Miklos Szeredi wrote: > > > > I would definitely prefer EEXIST logics dealt with in fs/namei.c - if nothing > > > else, it had been done wrong in too many instances... > > > > Okay. > > OK, so I'm taking 1, 5, 9, 10, 11, then add check to fs/namei.c:atomic_open(), > leaving the rest of gfs2 ones for gfs2 tree... Give me a few and I'll push > that. Pushed (head at d332c7). That should cover 9p and fuse as well, AFAICS. Do you have any problems with that variant?