From mboxrd@z Thu Jan 1 00:00:00 1970 From: Johannes Berg Subject: Re: [PATCH 1/1] net: hold sock reference while processing tx timestamps Date: Thu, 13 Oct 2011 10:54:43 +0200 Message-ID: <1318496083.5612.1.camel@jlt3.sipsolutions.net> References: <1318007501.3988.20.camel@jlt3.sipsolutions.net> <56185ca8a7dc0223031ca0f0996302cac1b497eb.1318444117.git.richard.cochran@omicron.at> (sfid-20111012_203706_058621_68EB8D6B) <1318447673.3933.47.camel@jlt3.sipsolutions.net> <1318449159.2644.9.camel@edumazet-laptop> (sfid-20111012_215313_649610_CEB1D4B5) Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: Richard Cochran , netdev@vger.kernel.org, David Miller To: Eric Dumazet Return-path: Received: from he.sipsolutions.net ([78.46.109.217]:34768 "EHLO sipsolutions.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753747Ab1JMIyt (ORCPT ); Thu, 13 Oct 2011 04:54:49 -0400 In-Reply-To: <1318449159.2644.9.camel@edumazet-laptop> (sfid-20111012_215313_649610_CEB1D4B5) Sender: netdev-owner@vger.kernel.org List-ID: On Wed, 2011-10-12 at 21:52 +0200, Eric Dumazet wrote: > > The other thing I was wondering -- what if we just set truesize to 1 (we > > don't have any truesize checks) and account the skb to the socket > > normally. Not really a good way either though. > > > > Changing truesize is not allowed here see below. Oh, good point. > > FWIW I just decided to do it the other way around in mac80211 -- keep > > the original SKB that's charged to the socket for the error queue, and > > use a clone to actually do the TX. > > Hmm, please take a look at IFF_SKB_TX_SHARING stuff, added in commit > d88733150 (net: add IFF_SKB_TX_SHARED flag to priv_flags) mac80211 got that flag cleared -- and right now I see no way to change that. 11ac/ad may require work in this area to be able to do this better, and I have a number of ideas on how to do that, but right now I don't see how we can do that. johannes