From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Wed, 19 Mar 2014 11:02:07 +0100 From: Linus =?utf-8?Q?L=C3=BCssing?= Message-ID: <20140319100207.GJ5208@Linus-Debian> References: <1395216532-31099-1-git-send-email-antonio@meshcoding.com> <1395216532-31099-2-git-send-email-antonio@meshcoding.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: <1395216532-31099-2-git-send-email-antonio@meshcoding.com> Subject: Re: [B.A.T.M.A.N.] [PATCHv2 next 02/11] batman-adv: don't mess up with the netdev refcounting if not needed Reply-To: The list for a Better Approach To Mobile Ad-hoc Networking List-Id: The list for a Better Approach To Mobile Ad-hoc Networking List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Antonio Quartulli Cc: b.a.t.m.a.n@lists.open-mesh.org On Wed, Mar 19, 2014 at 09:08:43AM +0100, Antonio Quartulli wrote: > Looking for an interface and increasing its refcounter only > to check if it really exists is not really useful and makes > the code look more complex without a real gain. Hm, I was intending to use it somewhere else as well with the next series of patches. See: http://git.open-mesh.org/batman-adv.git/commitdiff/bd774bd7e00bbe062cb0ab591d50be77e30a557a?hp=c9c3d53851f832b32188c9dd9a3dace96a33a0c3 But ok, since it is not in a yet scheduled patchset, I guess you're right, then it should better be introduced when it is needed. On another note, I think you're introducing a bug/memoryleak here for the compat-code case for netdev_master_upper_dev_get_rcu(). Cheers, Linus