From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1750941AbdIBArH (ORCPT ); Fri, 1 Sep 2017 20:47:07 -0400 Received: from mail-pf0-f169.google.com ([209.85.192.169]:34983 "EHLO mail-pf0-f169.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750794AbdIBArF (ORCPT ); Fri, 1 Sep 2017 20:47:05 -0400 X-Google-Smtp-Source: ADKCNb61dqlC6FWOM9GvL/WF3HnQSKtXd04WXcoGSqT22Qlu5rQhCzJCwkyE4ciaBbQXJJIhfVYYew== Date: Fri, 1 Sep 2017 17:47:01 -0700 From: Brian Norris To: Jeffy Chen Cc: linux-kernel@vger.kernel.org, dbasehore@chromium.org, dianders@chromium.org, Chanwoo Choi , Kyungmin Park , MyungJoo Ham , linux-pm@vger.kernel.org Subject: Re: [PATCH] CHROMIUM: devfreq: rk3399: Clear edev->dev drvdata before enabling dfi Message-ID: <20170902004659.GA105370@google.com> References: <20170901235237.18390-1-jeffy.chen@rock-chips.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20170901235237.18390-1-jeffy.chen@rock-chips.com> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sat, Sep 02, 2017 at 07:52:37AM +0800, Jeffy Chen wrote: > Currently we are using edev->dev drvdata to get rk3399-dmc data, but > it would be inited to edev in devfreq_event_add_edev. > > So we need to clear the edev->dev drvdata before enabling dfi, to > prevent dfi from getting the wrong rk3399-dmc data when the irq > triggered too early. Your description doesn't match your code. You say you're clearing evdev->dev driver data but... > Signed-off-by: Jeffy Chen > --- > > drivers/devfreq/rk3399_dmc.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/drivers/devfreq/rk3399_dmc.c b/drivers/devfreq/rk3399_dmc.c > index 1b89ebbad02c..12f9f03f349f 100644 > --- a/drivers/devfreq/rk3399_dmc.c > +++ b/drivers/devfreq/rk3399_dmc.c > @@ -429,6 +429,7 @@ static int rk3399_dmcfreq_probe(struct platform_device *pdev) > > rk3399_devfreq_dmc_profile.initial_freq = data->rate; > > + platform_set_drvdata(pdev, NULL); ...here you're only clearing the drvdata for the platform device. Is that a mistake? (Hint: that's not what you uploaded on the Chromium OS instance, where you presumably tested this.) And if you're really trying to do what your commit message says: We're having two different files fight over who owns the edev drvdata? That seems like a big no-no. We should work out who's the real owner of 'drvdata', and find some other solution for the others. Brian > data->devfreq = devm_devfreq_add_device(dev, > &rk3399_devfreq_dmc_profile, > "simple_ondemand", > -- > 2.11.0 > >