From mboxrd@z Thu Jan 1 00:00:00 1970 From: Cyrill Gorcunov Subject: Re: [patch 3/8] procfs: Add ability to plug in auxiliary fdinfo providers Date: Wed, 15 Aug 2012 01:56:16 +0400 Message-ID: <20120814215616.GC31043@moon> References: <20120814140342.354405844@openvz.org> <20120814140620.033884909@openvz.org> <20120814183142.GH23464@ZenIV.linux.org.uk> <20120814183558.GA1551@moon> <20120814195649.GB31043@moon> <20120814212721.GJ23464@ZenIV.linux.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org, Alexey Dobriyan , Andrew Morton , Pavel Emelyanov , James Bottomley , Matthew Helsley To: Al Viro Return-path: Content-Disposition: inline In-Reply-To: <20120814212721.GJ23464@ZenIV.linux.org.uk> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On Tue, Aug 14, 2012 at 10:27:21PM +0100, Al Viro wrote: > On Tue, Aug 14, 2012 at 11:56:49PM +0400, Cyrill Gorcunov wrote: > > Al, does the patch below looks better? If so I'll fix up the rest. > > > struct file_operations { > > struct module *owner; > > + struct seq_operations *fdinfo_ops; > > IDGI. Why on the earth do you need the whole iterator? All it takes > is show_fdinfo(struct seq_file *m, struct file *f); have ->show() in > your iterator (one going through the files) call that sucker for the > file we are trying to show. > > I think you are severely overdesigning that thing... Hmm, in very first versions I've been using one ->show method, but then I thought that this is not very correlate with seq-files idea where for each record show/next sequence is called. I'll update (this for sure will make code simplier, and I'll have to check for seq-file overflow after seq_printf call to not continue printing data for too long if buffer already out of space). Cyrill