Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752034AbdHASxW (ORCPT ); Tue, 1 Aug 2017 14:53:22 -0400 Received: from mail-wm0-f47.google.com ([74.125.82.47]:35985 "EHLO mail-wm0-f47.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751877AbdHASxU (ORCPT ); Tue, 1 Aug 2017 14:53:20 -0400 Subject: Re: [PATCH net-next 10/11] net: dsa: mv88e6xxx: remove EEE support To: Andrew Lunn Cc: Vivien Didelot , netdev@vger.kernel.org, linux-kernel@vger.kernel.org, kernel@savoirfairelinux.com, "David S. Miller" References: <20170731221719.16695-1-vivien.didelot@savoirfairelinux.com> <20170731221719.16695-11-vivien.didelot@savoirfairelinux.com> <20170801142843.GE23157@lunn.ch> <87r2wv2tci.fsf@weeman.i-did-not-set--mail-host-address--so-tickle-me> <20170801160628.GJ23157@lunn.ch> <17c26dee-c285-4dd5-1be0-286c0126d4db@gmail.com> <20170801172757.GL23157@lunn.ch> From: Florian Fainelli Message-ID: <9b9b2268-7409-fa37-f26f-3206a450c94a@gmail.com> Date: Tue, 1 Aug 2017 11:53:14 -0700 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.2.1 MIME-Version: 1.0 In-Reply-To: <20170801172757.GL23157@lunn.ch> Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 976 Lines: 24 On 08/01/2017 10:27 AM, Andrew Lunn wrote: >> If the MAC does not support EEE but the PHY does I think you can still >> allow EEE to be advertised and enabled, you just won't have the MAC be >> able to leverage the power savings that EEE brings. AFAICT this is still >> a valid mode whereby the PHY is put in a lower power mode, just not the >> whole transmit path (MAC + PHY). > > Hi Florian > > I read a couple of datasheets for a few phys doing EEE. Both said the > same, the MAC has to indicate to the PHY when low power should be > entered and existed. The PHY itself does not appear to do anything on > its own. Oh you are right, the LPI signal has to come from the MAC for the PHY to decide how to do the idle signaling, my bad. > > It would be good to read the standards about this. But i don't think > we should tell userspace EEE is enabled, if the PHY has it enabled, > but the MAC is not capable and hence EEE is not actually being used at > all. -- Florian