Return-Path: MIME-Version: 1.0 In-Reply-To: <20130203185107.0dcdfc4271d63c9e5403d697@studenti.unina.it> References: <1359908071-15024-1-git-send-email-ospite@studenti.unina.it> <1359908071-15024-3-git-send-email-ospite@studenti.unina.it> <20130203185107.0dcdfc4271d63c9e5403d697@studenti.unina.it> Date: Sun, 3 Feb 2013 14:27:26 -0400 Message-ID: Subject: Re: [PATCH BlueZ 2/2] device: use device_set_trusted() in set_trust() From: Anderson Lizardo To: Antonio Ospite Cc: linux-bluetooth@vger.kernel.org Content-Type: text/plain; charset=ISO-8859-1 Sender: linux-bluetooth-owner@vger.kernel.org List-ID: Hi Antonio, On Sun, Feb 3, 2013 at 1:51 PM, Antonio Ospite wrote: > The logic is equivalent to the previous one, as device_set_trusted() > returns without doing anything when (device->trusted == trusted), and > the previous implementation was already calling > g_dbus_pending_property_success() just before returning in any exit > path. Or was it meant to be g_dbus_pending_property_error() in the > first exit path? I don't know about that. My mistake. I incorrectly read the diff. And I think the _success() function is correct as it is. > > The previous set_trust() code didn't consider the !device case so we > should be OK if this function assumes the device is always defined. > > If the code is OK I can improve the commit message, yes. Having more text on the commit message is always better than less text :). But in this case, the commit is quite simple, so it is up to you/Johan/Marcel. Best Regards, -- Anderson Lizardo Instituto Nokia de Tecnologia - INdT Manaus - Brazil