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: <20210312165017.31829-1-frederic.danis@collabora.com>
 <CABBYNZ+fjE8x6Ba4FJvzb+RUoNgN9i8O8Vn4g0fhLJ+8CbYigg@mail.gmail.com> <49e20ff2-795c-12e9-db79-697b12725f12@collabora.com>
In-Reply-To: <49e20ff2-795c-12e9-db79-697b12725f12@collabora.com>
From:   Luiz Augusto von Dentz <luiz.dentz@gmail.com>
Date:   Fri, 12 Mar 2021 11:37:46 -0800
Message-ID: <CABBYNZKOW_v+KHz-jjrMzgUsBpk+axze8UL4BUp-xnpgk1mZqw@mail.gmail.com>
Subject: Re: [PATCH Bluez v2] adapter: Fix discovery trigger for 0 second delay
To:     =?UTF-8?B?RnLDqWTDqXJpYyBEYW5pcw==?= <frederic.danis@collabora.com>
Cc:     "linux-bluetooth@vger.kernel.org" <linux-bluetooth@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Hi Fr=C3=A9d=C3=A9ric,

On Fri, Mar 12, 2021 at 11:09 AM Fr=C3=A9d=C3=A9ric Danis
<frederic.danis@collabora.com> wrote:
>
> Hi Luiz,
>
> On 12/03/2021 19:28, Luiz Augusto von Dentz wrote:
> > Hi Fr=C3=A9d=C3=A9ric,
> >
> > On Fri, Mar 12, 2021 at 8:53 AM Fr=C3=A9d=C3=A9ric Danis
> > <frederic.danis@collabora.com> wrote:
> >> When calling `StartDiscovery` the effective start can take around 10 m=
s or
> >> up to 700 ms.
> >> g_timeout_add_seconds() call doesn't ensure the time for the first cal=
l of
> >> the timer if the delay is less or equal to 1 second.
> > Interesting, I always thought that 0 would be handle just as idle and
> > not round up to the next timeout.
> >
> >> ---
> >> v2: Fix issue founs by CI
> >>
> >>   src/adapter.c | 7 +++++++
> >>   1 file changed, 7 insertions(+)
> >>
> >> diff --git a/src/adapter.c b/src/adapter.c
> >> index cc0849f99..3078ce1a8 100644
> >> --- a/src/adapter.c
> >> +++ b/src/adapter.c
> >> @@ -1797,6 +1797,13 @@ static void trigger_start_discovery(struct btd_=
adapter *adapter, guint delay)
> >>          if (!btd_adapter_get_powered(adapter))
> >>                  return;
> >>
> >> +       if (!delay) {
> >> +               adapter->discovery_idle_timeout =3D g_idle_add(
> >> +                                               start_discovery_timeou=
t,
> >> +                                               adapter);
> >> +               return;
> >> +       }
> >> +
> >>          adapter->discovery_idle_timeout =3D g_timeout_add_seconds(del=
ay,
> >>                                          start_discovery_timeout, adap=
ter);
> > Maybe we should have a wrapper function for g_timeout_add_seconds
> > since I suspect there might be other instances of
> > g_timeout_add_seconds with 0 delay.
>
> Ok
> Is adding a timeout_add_seconds() function to src/shared/timeout.h the
> right place?

Yep, that should be the right place.

--=20
Luiz Augusto von Dentz