Received-SPF: pass (google.com: domain of linux-bluetooth-owner@vger.kernel.org designates 23.128.96.34 as permitted sender) client-ip=23.128.96.34;
MIME-Version: 1.0
References: <20231002153352.3163-1-iulia.tanasescu@nxp.com> <20231002153352.3163-4-iulia.tanasescu@nxp.com>
In-Reply-To: <20231002153352.3163-4-iulia.tanasescu@nxp.com>
From:   Luiz Augusto von Dentz <luiz.dentz@gmail.com>
Date:   Mon, 2 Oct 2023 14:30:40 -0700
Message-ID: <CABBYNZKiTS9herQBkMZge3CienN46_1TDcjdFUPf+yw05ESwtQ@mail.gmail.com>
Subject: Re: [PATCH v2 3/3] bap:Update transport acquire/release flow for
 bcast src
To:     Iulia Tanasescu <iulia.tanasescu@nxp.com>
Cc:     linux-bluetooth@vger.kernel.org, claudia.rosu@nxp.com,
        mihai-octavian.urzica@nxp.com, silviu.barbulescu@nxp.com,
        vlad.pruteanu@nxp.com, andrei.istodorescu@nxp.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Hi Iulia,

On Mon, Oct 2, 2023 at 8:34=E2=80=AFAM Iulia Tanasescu <iulia.tanasescu@nxp=
.com> wrote:
>
> From: Silviu Florian Barbulescu <silviu.barbulescu@nxp.com>
>
> Update transport acquire/release flow for BAP bcast source
>
> ---
>  profiles/audio/bap.c | 16 ++++++++++------
>  1 file changed, 10 insertions(+), 6 deletions(-)
>
> diff --git a/profiles/audio/bap.c b/profiles/audio/bap.c
> index 48a1a4f86..9a46b16ab 100644
> --- a/profiles/audio/bap.c
> +++ b/profiles/audio/bap.c
> @@ -1328,6 +1328,10 @@ static void iso_connect_cb(GIOChannel *chan, GErro=
r *err, gpointer user_data)
>         fd =3D g_io_channel_unix_get_fd(chan);
>
>         if (bt_bap_stream_set_io(stream, fd)) {
> +               if (bt_bap_stream_get_type(stream) =3D=3D
> +                                               BT_BAP_STREAM_TYPE_BCAST)=
 {
> +                       bt_bap_stream_start(stream, NULL, NULL);
> +               }

Perhaps it would be a better idea to have a dedicated callback for
broadcast in this case, so we don't mix the flows with unicast, since
the handling is quite different. In fact I was going to suggest we
change the broadcast code to live on bcaa driver rather than bap, so
we can move the handing on adapter_probe/adapter_remove there instead
of messing up with bap.

>                 g_io_channel_set_close_on_unref(chan, FALSE);
>                 return;
>         }
> @@ -1887,18 +1891,16 @@ static void bap_state(struct bt_bap_stream *strea=
m, uint8_t old_state,
>                 }
>                 break;
>         case BT_BAP_STREAM_STATE_QOS:
> -               bap_create_io(data, ep, stream, true);
> +               if (bt_bap_stream_get_type(stream) =3D=3D
> +                                       BT_BAP_STREAM_TYPE_UCAST) {
> +                       bap_create_io(data, ep, stream, true);
> +               }
>                 break;
>         case BT_BAP_STREAM_STATE_ENABLING:
>                 if (ep)
>                         bap_create_io(data, ep, stream, false);
>                 break;
>         case BT_BAP_STREAM_STATE_STREAMING:
> -               if (bt_bap_stream_get_type(stream) =3D=3D
> -                               BT_BAP_STREAM_TYPE_BCAST) {
> -                       if (ep)
> -                               bap_create_io(data, ep, stream, false);
> -               }
>                 break;
>         }
>  }
> @@ -2116,6 +2118,8 @@ static void bap_connecting(struct bt_bap_stream *st=
ream, bool state, int fd,
>
>                         ep->qos.bcast.big =3D qos.bcast.big;
>                         ep->qos.bcast.bis =3D qos.bcast.bis;
> +                       bt_bap_stream_config(ep->stream, &ep->qos,
> +                                       ep->caps, NULL, NULL);
>                 }
>
>                 DBG("stream %p fd %d: BIG 0x%02x BIS 0x%02x", stream, fd,
> --
> 2.39.2
>


--=20
Luiz Augusto von Dentz