Re: [PATCH] Bluetooth: hci_core: lookup pa sync need check BIG sync state

[Yang Li <yang.li@amlogic.com>]

Hi Luiz,
> [ EXTERNAL EMAIL ]
>
> Hi,
>
> On Tue, Jul 1, 2025 at 9:18 PM Yang Li via B4 Relay
> <devnull+yang.li.amlogic.com@kernel.org> wrote:
>> From: Yang Li <yang.li@amlogic.com>
>>
>> Ignore the big sync connections, we are looking for the PA
>> sync connection that was created as a result of the PA sync
>> established event.
> Were you seeing an issue with this, if you do please describe it and
> add the traces, debug logs, etc.

Since the PA sync connection is set to BT_CONNECTED in 
hci_le_big_sync_established_evt, if its status is BT_CONNECTED when 
hci_abort_conn_sync is called, hci_disconnect_sync() will be executed, 
which will cause the PA sync connection to be deleted.

int hci_abort_conn_sync(struct hci_dev *hdev, struct hci_conn *conn, u8 
reason)
{
...
     switch (conn->state) {
     case BT_CONNECTED:
     case BT_CONFIG:
         err = hci_disconnect_sync(hdev, conn, reason);
         break;
...

stack trace as below:

[   55.154495][0 T1966  d.] CPU: 0 PID: 1966 Comm: kworker/u9:0 Tainted: 
G           O       6.6.77 #104
[   55.155721][0 T1966  d.] Hardware name: Amlogic (DT)
[   55.156336][0 T1966  d.] Workqueue: hci0 hci_cmd_sync_work
[   55.157018][0 T1966  d.] Call trace:
[   55.157461][0 T1966  d.]  dump_backtrace+0x94/0xec
[   55.158056][0 T1966  d.]  show_stack+0x18/0x24
[   55.158607][0 T1966  d.]  dump_stack_lvl+0x48/0x60
[   55.159205][0 T1966  d.]  dump_stack+0x18/0x24
[   55.159756][0 T1966  d.]  hci_conn_del+0x1c/0x12c
[   55.160341][0 T1966  d.]  hci_conn_failed+0xdc/0x150
[   55.160958][0 T1966  d.]  hci_abort_conn_sync+0x204/0x388
[   55.161630][0 T1966  d.]  abort_conn_sync+0x58/0x80
[   55.162237][0 T1966  d.]  hci_cmd_sync_work+0x94/0x100
[   55.162877][0 T1966  d.]  process_one_work+0x168/0x444
[   55.163516][0 T1966  d.]  worker_thread+0x378/0x3f4
[   55.164122][0 T1966  d.]  kthread+0x108/0x10c
[   55.164664][0 T1966  d.]  ret_from_fork+0x10/0x20
[   55.165408][0 T1966  d.] hci0 hcon 000000004f36962c handle 3841 #PA 
sync connection


btmon trace:

< HCI Command: Disconnect (0x01|0x0006) plen 3             #75 [hci0] 
14.640630
         Handle: 3841
         Reason: Remote User Terminated Connection (0x13)
 > HCI Event: Command Status (0x0f) plen 4                  #76 [hci0] 
14.642103
       Disconnect (0x01|0x0006) ncmd 1
         Status: Invalid HCI Command Parameters (0x12)


So the current question is whether the PA sync connection, which is 
marked as BT_CONNECTED, really needs to be disconnected.
If it does need to be disconnected, then the PA sync terminate command 
must be executed.
However, in my opinion, the PA sync connection should not be disconnected.

>
>> Signed-off-by: Yang Li <yang.li@amlogic.com>
>> ---
>>   include/net/bluetooth/hci_core.h | 7 +++++++
>>   1 file changed, 7 insertions(+)
>>
>> diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
>> index 3ce1fb6f5822..646b0c5fd7a5 100644
>> --- a/include/net/bluetooth/hci_core.h
>> +++ b/include/net/bluetooth/hci_core.h
>> @@ -1400,6 +1400,13 @@ hci_conn_hash_lookup_pa_sync_handle(struct hci_dev *hdev, __u16 sync_handle)
>>                  if (c->type != BIS_LINK)
>>                          continue;
>>
>> +               /* Ignore the big sync connections, we are looking
>> +                * for the PA sync connection that was created as
>> +                * a result of the PA sync established event.
>> +                */
>> +               if (test_bit(HCI_CONN_BIG_SYNC, &c->flags))
>> +                       continue;
>> +
> hci_conn_hash_lookup_pa_sync_big_handle does:
>
>          if (c->type != BIS_LINK ||
>              !test_bit(HCI_CONN_PA_SYNC, &c->flags))
>
>>                  /* Ignore the listen hcon, we are looking
>>                   * for the child hcon that was created as
>>                   * a result of the PA sync established event.
>>
>> ---
>> base-commit: 3bc46213b81278f3a9df0324768e152de71eb9fe
>> change-id: 20250701-pa_sync-2fc7fc9f592c
>>
>> Best regards,
>> --
>> Yang Li <yang.li@amlogic.com>
>>
>>
>
> --
> Luiz Augusto von Dentz