Hi Luiz,
On Thu, Oct 27, 2022, at 20:59, Luiz Augusto von Dentz wrote:
> Hi Sven,
>
> On Thu, Oct 27, 2022 at 8:09 AM Sven Peter <sven@xxxxxxxxxxxxx> wrote:
>>
>> Broadcom 4378/4387 controllers found in Apple Silicon Macs claim to
>> support getting MWS Transport Layer Configuration,
>>
>> < HCI Command: Read Local Supported... (0x04|0x0002) plen 0
>> > HCI Event: Command Complete (0x0e) plen 68
>> Read Local Supported Commands (0x04|0x0002) ncmd 1
>> Status: Success (0x00)
>> [...]
>> Get MWS Transport Layer Configuration (Octet 30 - Bit 3)]
>> [...]
>>
>> , but then don't actually allow the required command:
>>
>> > HCI Event: Command Complete (0x0e) plen 15
>> Get MWS Transport Layer Configuration (0x05|0x000c) ncmd 1
>> Status: Command Disallowed (0x0c)
>> Number of transports: 0
>> Baud rate list: 0 entries
>> 00 00 00 00 00 00 00 00 00 00
>>
>> Signed-off-by: Sven Peter <sven@xxxxxxxxxxxxx>
>> ---
>> include/net/bluetooth/hci.h | 10 ++++++++++
>> net/bluetooth/hci_sync.c | 2 ++
>> 2 files changed, 12 insertions(+)
>>
>> diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h
>> index 8cd89948f961..110d6df1299b 100644
>> --- a/include/net/bluetooth/hci.h
>> +++ b/include/net/bluetooth/hci.h
>> @@ -273,6 +273,16 @@ enum {
>> * during the hdev->setup vendor callback.
>> */
>> HCI_QUIRK_BROKEN_EXT_SCAN,
>> +
>> + /*
>> + * When this quirk is set, the HCI_OP_GET_MWS_TRANSPORT_CONFIG command is
>> + * disabled. This is required for some Broadcom controllers which
>> + * erroneously claim to support MWS Transport Layer Configuration.
>> + *
>> + * This quirk can be set before hci_register_dev is called or
>> + * during the hdev->setup vendor callback.
>> + */
>> + HCI_QUIRK_BROKEN_MWS_TRANSPORT_CONFIG,
>> };
>>
>> /* HCI device flags */
>> diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
>> index 76c3107c9f91..91788d356748 100644
>> --- a/net/bluetooth/hci_sync.c
>> +++ b/net/bluetooth/hci_sync.c
>> @@ -4260,6 +4260,8 @@ static int hci_get_mws_transport_config_sync(struct hci_dev *hdev)
>> {
>> if (!(hdev->commands[30] & 0x08))
>> return 0;
>> + if (test_bit(HCI_QUIRK_BROKEN_MWS_TRANSPORT_CONFIG, &hdev->quirks))
>> + return 0;
>
> Let's add a macro that tests both the command and the quirk so we
> don't have to test them separately.
Sure, I'll add a macro for v5.
Best,
Sven
