Fwd: Bluetooth LE scan doesn't show device name

From: Bagas Sanjaya
Date: Tue Aug 08 2023 - 15:46:30 EST


Hi,

I notice a regression report on Bugzilla [1]. Quoting from it:

> Hi!
> At some point after kernel 5.9 I started having issues with LE device scanning.
> Here's how it used to work:
>
> $ bluetoothctl
> # power on
> # scan on
> ...
> [NEW] Device 68:71:DD:73:97:D5 Playfinity-2
>
> It successfully finds my device and it's name.
> On a newer kernel instead I'm getting no name:
>
> [NEW] Device 4D:18:19:A8:63:B5 4D-18-19-A8-63-B5
>
> Here's corresponding btmon logs. First from kernel 5.9.12 that can see device's name:
>
> ```5.9.12
>> HCI Event: LE Meta Event (0x3e) plen 33 #118 [hci0] 5.607028
> LE Extended Advertising Report (0x0d)
> Num reports: 1
> Entry 0
> Event type: 0x0013
> Props: 0x0013
> Connectable
> Scannable
> Use legacy advertising PDUs
> Data status: [0;32mComplete [0m
> Legacy PDU Type: ADV_IND (0x0013)
> Address type: Random (0x01)
> Address: 68:71:DD:73:97:D5 (Resolvable)
> Primary PHY: LE 1M
> Secondary PHY: No packets
> SID: no ADI field (0xff)
> TX power: 127 dBm
> RSSI: -54 dBm (0xca)
> Periodic advertising interval: 0.00 msec (0x0000)
> Direct address type: Public (0x00)
> Direct address: 00:00:00:00:00:00 (OUI 00-00-00)
> Data length: 0x07
> 02 01 06 03 02 f0 ff .......
> Flags: 0x06
> LE General Discoverable Mode
> BR/EDR Not Supported
> 16-bit Service UUIDs (partial): 1 entry
> Unknown (0xfff0)
>> HCI Event: LE Meta Event (0x3e) plen 49 #119 [hci0] 5.608029
> LE Extended Advertising Report (0x0d)
> Num reports: 1
> Entry 0
> Event type: 0x001b
> Props: 0x001b
> Connectable
> Scannable
> Scan response
> Use legacy advertising PDUs
> Data status: [0;32mComplete [0m
> Legacy PDU Type: SCAN_RSP to an ADV_SCAN_IND (0x001b)
> Address type: Random (0x01)
> Address: 68:71:DD:73:97:D5 (Resolvable)
> Primary PHY: LE 1M
> Secondary PHY: No packets
> SID: no ADI field (0xff)
> TX power: 127 dBm
> RSSI: -54 dBm (0xca)
> Periodic advertising interval: 0.00 msec (0x0000)
> Direct address type: Public (0x00)
> Direct address: 00:00:00:00:00:00 (OUI 00-00-00)
> Data length: 0x17
> 0d 09 50 6c 61 79 66 69 6e 69 74 79 2d 32 02 0a ..Playfinity-2..
> 00 05 12 50 00 68 00 ...P.h.
> Name (complete): Playfinity-2
> TX power: 0 dBm
> Peripheral Conn. Interval: 0x0050 - 0x0068
> ```
>
> And from 6.4.8:
>
> ```6.4.8
>> HCI Event: LE Meta Event (0x3e) plen 33 #130 [hci0] 9.180207
> LE Extended Advertising Report (0x0d)
> Num reports: 1
> Entry 0
> Event type: 0x0013
> Props: 0x0013
> Connectable
> Scannable
> Use legacy advertising PDUs
> Data status: �[0;32mComplete�[0m
> Legacy PDU Type: ADV_IND (0x0013)
> Address type: Random (0x01)
> Address: 4D:18:19:A8:63:B5 (Resolvable)
> Primary PHY: LE 1M
> Secondary PHY: No packets
> SID: no ADI field (0xff)
> TX power: 127 dBm
> RSSI: -53 dBm (0xcb)
> Periodic advertising interval: 0.00 msec (0x0000)
> Direct address type: Public (0x00)
> Direct address: 00:00:00:00:00:00 (OUI 00-00-00)
> Data length: 0x07
> 02 01 06 03 02 f0 ff .......
> Flags: 0x06
> LE General Discoverable Mode
> BR/EDR Not Supported
> 16-bit Service UUIDs (partial): 1 entry
> Unknown (0xfff0)
> ```
>
> I've tried compiling 5.9.12 to see if I can bissect, but it fails to compile with gcc12..
>
> Is it expected that newer kernels can't get device name? Perhaps some additional action is needed fetch it?
>
> Thanks!

See Bugzilla for the full thread.

TatriX: Can you also attach dmesg output to your Bugzilla report?
Since you also have kernel build problem, can you also attach build
log (``make 2>&1 | tee build.log``)?

Anyway, I'm adding this regression to be tracked by regzbot:

#regzbot introduced: v5.9..v6.4 https://bugzilla.kernel.org/show_bug.cgi?id=217773

Thanks.

[1]: https://bugzilla.kernel.org/show_bug.cgi?id=217773

--
An old man doll... just what I always wanted! - Clara