Return-Path: <cyrus@holtmann.org>
Date: Mon, 17 Oct 2011 10:31:24 +0300
From: Andrei Emeltchenko <andrei.emeltchenko.news@gmail.com>
To: Mat Martineau <mathewm@codeaurora.org>
Cc: Gustavo Padovan <padovan@profusion.mobi>,
	linux-bluetooth@vger.kernel.org, pkrystad@codeaurora.org
Subject: Re: [PATCH 9/9] Bluetooth: Add signal handlers for channel moves
Message-ID: <20111017073122.GC6724@aemeltch-MOBL1>
References: <1318543247-27130-1-git-send-email-mathewm@codeaurora.org>
 <1318543247-27130-10-git-send-email-mathewm@codeaurora.org>
 <20111014184650.GD30989@joana>
 <alpine.DEB.2.02.1110141618420.9993@mathewm-linux>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
In-Reply-To: <alpine.DEB.2.02.1110141618420.9993@mathewm-linux>
Sender: linux-bluetooth-owner@vger.kernel.org
List-ID: <linux-bluetooth.vger.kernel.org>

Hi Mat,

On Fri, Oct 14, 2011 at 04:19:53PM -0700, Mat Martineau wrote:
> >>AMP channels can be moved between BR/EDR and AMP controllers using a
> >>sequence of signals. Every attempted channel move involves a series of
> >>four signals:
> >>
> >>   Move Initiator                 Move Responder
> >>        |                                 |
> >>        |       Move Channel Request      |
> >>        |  ---------------------------->  |
> >>        |                                 |
> >>        |       Move Channel Response     |
> >>        |  <----------------------------  |
> >>        |                                 |
> >>        |       Move Channel Confirm      |
> >>        |  ---------------------------->  |
> >>        |                                 |
> >>        |  Move Channel Confirm Response  |
> >>	|  <----------------------------  |
> >>
> >>All four signals are sent even if the move fails.
> >>
> >>Signed-off-by: Mat Martineau <mathewm@codeaurora.org>
> >>---
> >> net/bluetooth/l2cap_core.c |  123 ++++++++++++++++++++++++++++++++++++++++++++
> >> 1 files changed, 123 insertions(+), 0 deletions(-)
> >>
> >>diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c
> >>index e38258b..ba64bab 100644
> >>--- a/net/bluetooth/l2cap_core.c
> >>+++ b/net/bluetooth/l2cap_core.c
> >>@@ -2957,6 +2957,112 @@ static inline int l2cap_create_channel_rsp(struct l2cap_conn *conn,
> >> 	return l2cap_connect_rsp(conn, cmd, data);
> >> }
> >>
> >>+static void l2cap_send_move_chan_rsp(struct l2cap_conn *conn, u8 ident,
> >>+				u16 icid, u16 result)
> >>+{
> >>+	struct l2cap_move_chan_rsp rsp;
> >>+
> >>+	BT_DBG("icid %d, result %d", (int) icid, (int) result);

BTW: why do you need type conversion here and below?

Best regards 
Andrei Emeltchenko 

> >>+
> >>+	rsp.icid = cpu_to_le16(icid);
> >>+	rsp.result = cpu_to_le16(result);
> >>+
> >>+	l2cap_send_cmd(conn, ident, L2CAP_MOVE_CHAN_RSP, sizeof(rsp), &rsp);
> >>+}
> >>+
> >>+static void l2cap_send_move_chan_cfm(struct l2cap_conn *conn,
> >>+				struct l2cap_chan *chan, u16 icid, u16 result)
> >>+{
> >>+	struct l2cap_move_chan_cfm cfm;
> >>+	u8 ident;
> >>+
> >>+	BT_DBG("icid %d, result %d", (int) icid, (int) result);
> >>+
> >>+	ident = l2cap_get_ident(conn);
> >>+	if (chan)
> >>+		chan->ident = ident;
> >>+
> >>+	cfm.icid = cpu_to_le16(icid);
> >>+	cfm.result = cpu_to_le16(result);
> >>+
> >>+	l2cap_send_cmd(conn, ident, L2CAP_MOVE_CHAN_CFM, sizeof(cfm), &cfm);
> >>+}
> >>+
> >>+static void l2cap_send_move_chan_cfm_rsp(struct l2cap_conn *conn, u8 ident,
> >>+					u16 icid)
> >>+{
> >>+	struct l2cap_move_chan_cfm_rsp rsp;
> >>+
> >>+	BT_DBG("icid %d", (int) icid);
> >>+
> >>+	rsp.icid = cpu_to_le16(icid);
> >>+	l2cap_send_cmd(conn, ident, L2CAP_MOVE_CHAN_CFM_RSP, sizeof(rsp), &rsp);
> >>+}
> >>+
> >>+static inline int l2cap_move_channel_req(struct l2cap_conn *conn,
> >>+					struct l2cap_cmd_hdr *cmd, u8 *data)
> >>+{
> >>+	struct l2cap_move_chan_req *req = (struct l2cap_move_chan_req *) data;
> >>+	u16 icid = 0;
> >>+	u16 result = L2CAP_MOVE_CHAN_REFUSED_NOT_ALLOWED;
> >>+
> >>+	icid = le16_to_cpu(req->icid);
> >>+
> >>+	BT_DBG("icid %d, dest_amp_id %d", (int) icid, (int) req->dest_amp_id);
> >>+
> >>+	/* Placeholder: Always refuse */
> >>+	l2cap_send_move_chan_rsp(conn, cmd->ident, icid, result);
> >>+
> >>+	return 0;
> >>+}
> >
> >It's a good idea protect the move channel request with enable_hs.
> 
> Yes, I'll require enable_hs to have the move channel do anything.
> When enable_hs is false, it seems like it should return a
> L2CAP_MOVE_CHAN_REFUSED_NOT_ALLOWED response like it does now.