[PATCH BlueZ 0/1] policy: fix HSP/HFP reconnection

[PATCH BlueZ 0/1] policy: fix HSP/HFP reconnection

[Yao Wei (魏銘廷)]

This patch was regarding to https://github.com/bluez/bluez/issues/1069.

The previous patch did not fix the issue, as I was testing it with Apple
AirPods Pro 2 and it automatically reconnects HSP/HFP after A2DP is
connected for about 25 seconds, and I assume the patch worked for me.
Some other devices I tested did not exhibit such behavior (Jabra Speak2
55 UC).  This patch was tested against both these devices, and both
cases with NVIDIA GPU enabled and disabled.

By removing the code checking whether A2DP service is initiator for
directly starting HSP/HFP service, it will now only check whether
HSP/HFP is connected, and start the service if not.

The previous patch also has an oversight of not falling back to get HSP
service if HFP is not available.  This patch fixed the issue.

Yao Wei (魏銘廷) (1):
  policy: fix HSP/HFP reconnection

 plugins/policy.c | 18 ++++++++----------
 1 file changed, 8 insertions(+), 10 deletions(-)

-- 
2.49.0

[PATCH BlueZ 1/1] policy: fix HSP/HFP reconnection

[Yao Wei (魏銘廷)]

This patch removes checking whether A2DP service initiates the
connection starting HSP/HFP connection, instead just start timer
when HSP/HFP is not connected, also move order so that setting
HSP/HFP connection timer is not blocked by AVRCP connection.

The previous patch also did not get HSP service if HFP is not
available.  This patch adds it in.
---
 plugins/policy.c | 18 ++++++++----------
 1 file changed, 8 insertions(+), 10 deletions(-)

diff --git a/plugins/policy.c b/plugins/policy.c
index 360d1359f..0e533ba1f 100644
--- a/plugins/policy.c
+++ b/plugins/policy.c
@@ -312,6 +312,14 @@ static void sink_cb(struct btd_service *service, btd_service_state_t old_state,
 			data->sink_timer = 0;
 		}
 
+		/* Try connecting HSP/HFP if it is not connected */
+		hs = btd_device_get_service(dev, HFP_HS_UUID);
+		if (hs == NULL)
+			hs = btd_device_get_service(data->dev, HSP_HS_UUID);
+		if (hs && btd_service_get_state(hs) !=
+						BTD_SERVICE_STATE_CONNECTED)
+			policy_set_hs_timer(data);
+
 		/* Check if service initiate the connection then proceed
 		 * immediatelly otherwise set timer
 		 */
@@ -321,16 +329,6 @@ static void sink_cb(struct btd_service *service, btd_service_state_t old_state,
 						BTD_SERVICE_STATE_CONNECTED)
 			policy_set_ct_timer(data, CONTROL_CONNECT_TIMEOUT);
 
-		/* Also try connecting HSP/HFP if it is not connected */
-		hs = btd_device_get_service(dev, HFP_HS_UUID);
-		if (hs) {
-			if (btd_service_is_initiator(service))
-				policy_connect(data, hs);
-			else if (btd_service_get_state(hs) !=
-						BTD_SERVICE_STATE_CONNECTED)
-				policy_set_hs_timer(data);
-		}
-
 		break;
 	case BTD_SERVICE_STATE_DISCONNECTING:
 		break;
-- 
2.49.0

RE: policy: fix HSP/HFP reconnection

[bluez.test.bot]

[-- Attachment #1: Type: text/plain, Size: 1261 bytes --]

This is automated email and please do not reply to this email!

Dear submitter,

Thank you for submitting the patches to the linux bluetooth mailing list.
This is a CI test results with your patch series:
PW Link:https://patchwork.kernel.org/project/bluetooth/list/?series=948733

---Test result---

Test Summary:
CheckPatch                    PENDING   0.33 seconds
GitLint                       PENDING   0.40 seconds
BuildEll                      PASS      20.84 seconds
BluezMake                     PASS      1562.68 seconds
MakeCheck                     PASS      12.91 seconds
MakeDistcheck                 PASS      161.21 seconds
CheckValgrind                 PASS      217.39 seconds
CheckSmatch                   PASS      289.96 seconds
bluezmakeextell               PASS      100.81 seconds
IncrementalBuild              PENDING   0.36 seconds
ScanBuild                     PASS      893.38 seconds

Details
##############################
Test: CheckPatch - PENDING
Desc: Run checkpatch.pl script
Output:

##############################
Test: GitLint - PENDING
Desc: Run gitlint
Output:

##############################
Test: IncrementalBuild - PENDING
Desc: Incremental build with the patches in the series
Output:



---
Regards,
Linux Bluetooth

Re: [PATCH BlueZ 0/1] policy: fix HSP/HFP reconnection

[patchwork-bot+bluetooth]

Hello:

This patch was applied to bluetooth/bluez.git (master)
by Luiz Augusto von Dentz <luiz.von.dentz@intel.com>:

On Tue,  1 Apr 2025 10:02:20 +0800 you wrote:
> This patch was regarding to https://github.com/bluez/bluez/issues/1069.
> 
> The previous patch did not fix the issue, as I was testing it with Apple
> AirPods Pro 2 and it automatically reconnects HSP/HFP after A2DP is
> connected for about 25 seconds, and I assume the patch worked for me.
> Some other devices I tested did not exhibit such behavior (Jabra Speak2
> 55 UC).  This patch was tested against both these devices, and both
> cases with NVIDIA GPU enabled and disabled.
> 
> [...]

Here is the summary with links:
  - [BlueZ,1/1] policy: fix HSP/HFP reconnection
    https://git.kernel.org/pub/scm/bluetooth/bluez.git/?id=0b3d49f4e030

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html