Received-SPF: pass (google.com: domain of linux-bluetooth-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20200501144037.1684-1-alainm@chromium.org> <4195BFB9-1586-435F-8FC2-ED215959A591@holtmann.org>
In-Reply-To: <4195BFB9-1586-435F-8FC2-ED215959A591@holtmann.org>
From:   Alain Michaud <alainmichaud@google.com>
Date:   Fri, 1 May 2020 13:17:45 -0400
Message-ID: <CALWDO_W4AquYjw5LZR+pbFnPEB=shYKDH7_JDGHexpsXUnC7aA@mail.gmail.com>
Subject: Re: [BlueZ PATCH v1] shared/gatt-client:Ignore orphaned characteristics
To:     Marcel Holtmann <marcel@holtmann.org>
Cc:     Alain Michaud <alainm@chromium.org>,
        BlueZ <linux-bluetooth@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-bluetooth-owner@vger.kernel.org
Precedence: bulk

Hi Marcel,


On Fri, May 1, 2020 at 12:26 PM Marcel Holtmann <marcel@holtmann.org> wrote:
>
> Hi Alain,
>
> > The gatt discovery proceedure simplification to discover all
> > characteristics at once has exposed a device side issue where some
> > device implementation reports orphaned characteristics.  While this
> > technically shouldn't be allowed, some instances of this were observed
> > (namely on some Android phones).
> >
> > The fix is to simply skip over orphaned characteristics and continue
> > with everything else that is valid.
> >
> > This has been tested as part of the Android CTS tests against an
> > affected platform and was confirmed to have worked around the issue.
> > ---
> >
> > src/shared/gatt-client.c | 5 ++++-
> > 1 file changed, 4 insertions(+), 1 deletion(-)
> >
> > diff --git a/src/shared/gatt-client.c b/src/shared/gatt-client.c
> > index 963ad619f..d604c77a3 100644
> > --- a/src/shared/gatt-client.c
> > +++ b/src/shared/gatt-client.c
> > @@ -632,7 +632,10 @@ static bool discover_descs(struct discovery_op *op, bool *discovering)
> >                       util_debug(client->debug_callback, client->debug_data,
> >                               "Failed to insert characteristic at 0x%04x",
> >                               chrc_data->value_handle);
> > -                     goto failed;
> > +
> > +                     /* Skip over invalid characteristic */
> > +                     free(chrc_data);
> > +                     continue;
> >               }
>
> should we add a warning here?
Is there a good way other than the util_debug mechanism to write  a warning?

>
> And I think it would be good to have a bit more verbose comment why this is done this way instead of aborting.
Ack, will be added in V2.

>
> Regards
>
> Marcel
>