that could be a reason and a potential fix, we remove the temp flag in
dev_connect and pair_device very early on, but i suspect changing that
would potentially have bigger impact and needs more due diligence and
testing.
On Mon, Aug 17, 2020 at 4:24 PM Luiz Augusto von Dentz
<luiz.dentz@xxxxxxxxx> wrote:
>
> Hi Yu Liu,
>
> On Mon, Aug 17, 2020 at 4:04 PM Yu Liu <yudiliu@xxxxxxxxxx> wrote:
> >
> > This caused the device hanging around as a discovered device forever
> > even if it is turned off or not in present.
> >
> > Reviewed-by: sonnysasaka@xxxxxxxxxxxx
> >
> > Signed-off-by: Yu Liu <yudiliu@xxxxxxxxxx>
> > ---
> >
> > Changes in v2:
> > - Fix title length and format issue
> >
> > Changes in v1:
> > - Initial change
> >
> > src/device.c | 6 ++++++
> > 1 file changed, 6 insertions(+)
> >
> > diff --git a/src/device.c b/src/device.c
> > index bb8e07e8f..93e71850c 100644
> > --- a/src/device.c
> > +++ b/src/device.c
> > @@ -6008,6 +6008,12 @@ void device_bonding_complete(struct btd_device *device, uint8_t bdaddr_type,
> >
> > if (status) {
> > device_cancel_authentication(device, TRUE);
> > +
> > + // Put the device back to the temporary state so that it will be
> > + // treated as a newly discovered device.
>
> Use C style comments /* */ above.
>
> > + if (!device_is_paired(device, bdaddr_type))
> > + btd_device_set_temporary(device, true);
>
> Hmm, are we perhaps removing the temporary flag too early? Or this is
> a result of calling Connect which clears the temporary flag? Then
> perhaps we should at least if the upper layer has marked the device as
> trusted as it should indicate the device object should be kept even if
> not paired.
>
> > device_bonding_failed(device, status);
> > return;
> > }
> > --
> > 2.28.0.220.ged08abb693-goog
> >
>
>
> --
> Luiz Augusto von Dentz
