Return-Path: MIME-Version: 1.0 In-Reply-To: References: <1421465191-1494-1-git-send-email-armansito@chromium.org> Date: Wed, 21 Jan 2015 22:28:38 +0200 Message-ID: Subject: Re: [PATCH BlueZ 0/2] core: device: GATT fixes From: Luiz Augusto von Dentz To: Arman Uguray Cc: BlueZ development Content-Type: text/plain; charset=UTF-8 Sender: linux-bluetooth-owner@vger.kernel.org List-ID: Hi Arman, On Tue, Jan 20, 2015 at 10:16 PM, Arman Uguray wrote: >> On Fri, Jan 16, 2015 at 7:26 PM, Arman Uguray wrote: >> This patch fixes the following: >> >> 1. Bugs in GATT service browsing code-path that caused GAttrib-based >> profiles to not work and caused D-Bus calls to "Pair" and/or "Connect" >> to hang. >> >> 2. Reintroduced invalid service removal. This patch is the similar as >> the one I submitted previously: it splits the UUID list into SDP and >> GATT based lists and introduces new storage entries, except this patch >> also includes a very simple migration mechanism from the existing >> storage format to the new one. >> >> The migration works by loading the UUIDs from the "Services" field and >> delaying profile probing until connection establishment. Once a >> connection is established and service discovery is performed, we >> simply store the UUIDs in their new grouping and probe the profiles >> and remove the previous entries. The legacy entries act as temporary >> data that's fed into the Device1.UUIDs property but everything else >> acts as if no services were discovered until connection. >> >> Arman Uguray (2): >> core: device: Fix bug in device_browse_gatt >> core: device: Fix broken GATT UUID management >> >> plugins/sixaxis.c | 2 +- >> src/adapter.c | 8 +- >> src/device.c | 410 +++++++++++++++++++++++++++++++++++++++--------------- >> src/device.h | 6 +- >> 4 files changed, 307 insertions(+), 119 deletions(-) >> >> -- >> 2.2.0.rc0.207.ga3a616c >> This needs rebasing, first patch Ive actually fix part of the problem but we still need to fix the part where it does not probe without browse, for the second patch I would skip for now since we anyway don't remove in case of SDP and it might have corner cases when for example an error happens during discovering this might leads to removing drivers. -- Luiz Augusto von Dentz