Skip to contentSkip to conversation

Bluetooth: Move BR/EDR default events behind its features

PATCH In review Marcel Holtmann 2 participants 4 messages Nov 1, 2015 – Nov 4, 2015

Subsystems (2)
lore
Share & tools
lore

Marcel Holtmann

author maintainer
#0 lore raw

PATCH 2/4 Bluetooth: Move BR/EDR default events behind its features

There are some BR/EDR default events for Bluetooth 1.2 or later controllers that are not conditional on their features being present.

---
 net/bluetooth/hci_core.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)
net/bluetooth/hci_core.c +8 -4
index 62edbf1b114e..db423657935a 100644
@@ -531,10 +531,6 @@ static void hci_setup_event_mask(struct hci_request *req)
 
 	if (lmp_bredr_capable(hdev)) {
 		events[4] |= 0x01; /* Flow Specification Complete */
-		events[4] |= 0x02; /* Inquiry Result with RSSI */
-		events[4] |= 0x04; /* Read Remote Extended Features Complete */
-		events[5] |= 0x08; /* Synchronous Connection Complete */
-		events[5] |= 0x10; /* Synchronous Connection Changed */
 	} else {
 		/* Use a different default for LE-only devices */
 		memset(events, 0, sizeof(events));
@@ -555,6 +551,14 @@ static void hci_setup_event_mask(struct hci_request *req)
 	if (lmp_inq_rssi_capable(hdev))
 		events[4] |= 0x02; /* Inquiry Result with RSSI */
 
+	if (lmp_ext_feat_capable(hdev))
+		events[4] |= 0x04; /* Read Remote Extended Features Complete */
+
+	if (lmp_esco_capable(hdev)) {
+		events[5] |= 0x08; /* Synchronous Connection Complete */
+		events[5] |= 0x10; /* Synchronous Connection Changed */
+	}
+
 	if (lmp_sniffsubr_capable(hdev))
 		events[5] |= 0x20; /* Sniff Subrating */
 
Signature · 1 line
2.4.3
#1 lore raw

Hi Marcel,

Quoted 7 lines · Marcel Holtmann

There are some BR/EDR default events for Bluetooth 1.2 or later controllers that are not conditional on their features being present.

---
 net/bluetooth/hci_core.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

I realize I already applied this one, but:

net/bluetooth/hci_core.c static void hci_setup_event_mask(struct hci_request *req) · line 560

Quoted 4 lines
net/bluetooth/hci_core.c +4 -0
+	if (lmp_esco_capable(hdev)) {
+		events[5] |= 0x08; /* Synchronous Connection Complete */
+		events[5] |= 0x10; /* Synchronous Connection Changed */
+	}

Synchronous Connection Complete is relevant also for non-ESCO controllers, isn't it?

Johan

author maintainer
#2 lore raw

Hi Johan,

Quoted 10 lines
Quoted 7 lines

There are some BR/EDR default events for Bluetooth 1.2 or later controllers that are not conditional on their features being present.

---
net/bluetooth/hci_core.c | 12 ++++++++----

1 file changed, 8 insertions(+), 4 deletions(-)

I realize I already applied this one, but:

Quoted 7 lines

net/bluetooth/hci_core.c static void hci_setup_event_mask(struct hci_request *req) · line 560

Quoted 4 lines
net/bluetooth/hci_core.c +4 -0
+	if (lmp_esco_capable(hdev)) {
+		events[5] |= 0x08; /* Synchronous Connection Complete */
+		events[5] |= 0x10; /* Synchronous Connection Changed */
+	}

Synchronous Connection Complete is relevant also for non-ESCO controllers, isn't it?

that is true, but if you look at hci_conn.c and sco.c, then you see we use the same logic there to decide if we use Add SCO or Setup Sync Connection. So this just brings that in sync.

Regards

Marcel

#3 lore raw

Hi Marcel,

Quoted 20 lines · Marcel Holtmann

Hi Johan,

Quoted 10 lines
Quoted 7 lines

There are some BR/EDR default events for Bluetooth 1.2 or later controllers that are not conditional on their features being present.

---
net/bluetooth/hci_core.c | 12 ++++++++----

1 file changed, 8 insertions(+), 4 deletions(-)

I realize I already applied this one, but:

Quoted 7 lines

net/bluetooth/hci_core.c static void hci_setup_event_mask(struct hci_request *req) · line 560

Quoted 4 lines
net/bluetooth/hci_core.c +4 -0
+	if (lmp_esco_capable(hdev)) {
+		events[5] |= 0x08; /* Synchronous Connection Complete */
+		events[5] |= 0x10; /* Synchronous Connection Changed */
+	}

Synchronous Connection Complete is relevant also for non-ESCO controllers, isn't it?

Quoted 3 lines

that is true, but if you look at hci_conn.c and sco.c, then you see we use the same logic there to decide if we use Add SCO or Setup Sync Connection. So this just brings that in sync.

Understood, it took some time for me to figure out from the core spec exactly under which conditions Conn Complete or Sync Conn Complete will happen.

Johan