From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx1.redhat.com ([209.132.183.28]:63011 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751006AbcGYNPc (ORCPT ); Mon, 25 Jul 2016 09:15:32 -0400 Date: Mon, 25 Jul 2016 09:15:30 -0400 From: Aristeu Rozanski Subject: Re: [PATCH 1/3] mac802154: don't warn on unsupported frames Message-ID: <20160725131530.GE30145@redhat.com> References: <1469207896-26481-1-git-send-email-arozansk@redhat.com> <78726941-7d65-1ece-e203-f1507597978a@pengutronix.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <78726941-7d65-1ece-e203-f1507597978a@pengutronix.de> Sender: linux-wpan-owner@vger.kernel.org List-ID: To: Alexander Aring Cc: linux-wpan@vger.kernel.org, Stefan Schmidt , Jukka Rissanen Hi Alexander, On Sat, Jul 23, 2016 at 02:46:28PM +0200, Alexander Aring wrote: > agree, but this patch introduce a different stats handling right now. :-) sorry, should have split both parts in different patches. > This should be moved into ieee802154_deliver_skb, right before calling > netif_rx(...); > > The reason is because we have for "stats.rx_bytes" and > "stats.rx_packets" a very clean definition what it means. It means, the > frames hit a state where it is ready to put it into the packet-layer, > this is what netif_rx is doing. > > It's okay for me to add support for rx_frame_errors and also rx_dropped > stuff. But with a very big warning that the meaning of such stats will > be touched again later if we found some strategie where we we define > what each stats attribute means. Also there exists subsystems which > simple ignore these stats, because at userspace side you need always to > lookup what the definition means and when we run such stats counters and > when not and some subsystems don't care about that. Fair enough, I'll redo the patch just removing the warning. Thanks! -- Aristeu