| Message ID | 20250224205439.18672-1-nishiyama.pedro@gmail.com |
|---|---|
| Headers | show |
| Series | Bluetooth: btusb: Fix regression in the initialization of fake Bluetooth controllers | expand |
Hi Luiz, On Mon, Feb 24, 2025 at 6:36 PM Luiz Augusto von Dentz <luiz.dentz@gmail.com> wrote: > > Hi Pedro, > > On Mon, Feb 24, 2025 at 3:54 PM Pedro Nishiyama > <nishiyama.pedro@gmail.com> wrote: > > > > This adds quirks for broken READ_VOICE_SETTING and READ_PAGE_SCAN_TYPE. > > > > Signed-off-by: Pedro Nishiyama <nishiyama.pedro@gmail.com> > > --- > > include/net/bluetooth/hci.h | 16 ++++++++++++++++ > > net/bluetooth/hci_sync.c | 6 ++++++ > > 2 files changed, 22 insertions(+) > > > > diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h > > index 0d51970d809f..b99818df8ee7 100644 > > --- a/include/net/bluetooth/hci.h > > +++ b/include/net/bluetooth/hci.h > > @@ -354,6 +354,22 @@ enum { > > * during the hdev->setup vendor callback. > > */ > > HCI_QUIRK_FIXUP_LE_EXT_ADV_REPORT_PHY, > > + > > + /* When this quirk is set, the HCI_OP_READ_VOICE_SETTING command is > > + * skipped. This is required for a subset of the CSR controller clones > > + * which erroneously claim to support it. > > + * > > + * This quirk must be set before hci_register_dev is called. > > + */ > > + HCI_QUIRK_BROKEN_READ_VOICE_SETTING, > > Lets split this in 2 parts, one for voice setting and another for page > scan type. > > > + /* When this quirk is set, the HCI_OP_READ_PAGE_SCAN_TYPE command is > > + * skipped. This is required for a subset of the CSR controller clones > > + * which erroneously claim to support it. > > + * > > + * This quirk must be set before hci_register_dev is called. > > + */ > > + HCI_QUIRK_BROKEN_READ_PAGE_SCAN_TYPE, > > }; > > > > /* HCI device flags */ > > diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c > > index dd770ef5ec36..e76012956020 100644 > > --- a/net/bluetooth/hci_sync.c > > +++ b/net/bluetooth/hci_sync.c > > @@ -3696,6 +3696,9 @@ static int hci_read_local_name_sync(struct hci_dev *hdev) > > /* Read Voice Setting */ > > static int hci_read_voice_setting_sync(struct hci_dev *hdev) > > { > > + if (test_bit(HCI_QUIRK_BROKEN_READ_VOICE_SETTING, &hdev->quirks)) > > + return 0; > > While at it I'd add the checking for the bit as well, that said > perhaps we need to disable SCO link if voice settings cannot be read, > is the controller able to create SCO connections? I did some tests on the version before the regression and with the patches. It is possible to create an SCO connection, but the connection is not usable. Before the regression there is no sound, and with the patches, dmesg is spammed with "Bluetooth: hci0: corrupted SCO packet". > > > return __hci_cmd_sync_status(hdev, HCI_OP_READ_VOICE_SETTING, > > 0, NULL, HCI_CMD_TIMEOUT); > > } > > @@ -4132,6 +4135,9 @@ static int hci_read_page_scan_type_sync(struct hci_dev *hdev) > > if (!(hdev->commands[13] & 0x01)) > > return 0; > > > > + if (test_bit(HCI_QUIRK_BROKEN_READ_PAGE_SCAN_TYPE, &hdev->quirks)) > > + return 0; > > + > > return __hci_cmd_sync_status(hdev, HCI_OP_READ_PAGE_SCAN_TYPE, > > 0, NULL, HCI_CMD_TIMEOUT); > > } > > -- > > 2.48.1 > > > > > -- > Luiz Augusto von Dentz -- Pedro Nishiyama
Hello, On Tue, Feb 25, 2025 at 12:55 PM Luiz Augusto von Dentz <luiz.dentz@gmail.com> wrote: > > Hi Pedro, > > On Mon, Feb 24, 2025 at 7:27 PM Pedro Nishiyama > <nishiyama.pedro@gmail.com> wrote: > > > > Hi Luiz, > > > > On Mon, Feb 24, 2025 at 6:36 PM Luiz Augusto von Dentz > > <luiz.dentz@gmail.com> wrote: > > > > > > Hi Pedro, > > > > > > On Mon, Feb 24, 2025 at 3:54 PM Pedro Nishiyama > > > <nishiyama.pedro@gmail.com> wrote: > > > > > > > > This adds quirks for broken READ_VOICE_SETTING and READ_PAGE_SCAN_TYPE. > > > > > > > > Signed-off-by: Pedro Nishiyama <nishiyama.pedro@gmail.com> > > > > --- > > > > include/net/bluetooth/hci.h | 16 ++++++++++++++++ > > > > net/bluetooth/hci_sync.c | 6 ++++++ > > > > 2 files changed, 22 insertions(+) > > > > > > > > diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h > > > > index 0d51970d809f..b99818df8ee7 100644 > > > > --- a/include/net/bluetooth/hci.h > > > > +++ b/include/net/bluetooth/hci.h > > > > @@ -354,6 +354,22 @@ enum { > > > > * during the hdev->setup vendor callback. > > > > */ > > > > HCI_QUIRK_FIXUP_LE_EXT_ADV_REPORT_PHY, > > > > + > > > > + /* When this quirk is set, the HCI_OP_READ_VOICE_SETTING command is > > > > + * skipped. This is required for a subset of the CSR controller clones > > > > + * which erroneously claim to support it. > > > > + * > > > > + * This quirk must be set before hci_register_dev is called. > > > > + */ > > > > + HCI_QUIRK_BROKEN_READ_VOICE_SETTING, > > > > > > Lets split this in 2 parts, one for voice setting and another for page > > > scan type. > > > > > > > + /* When this quirk is set, the HCI_OP_READ_PAGE_SCAN_TYPE command is > > > > + * skipped. This is required for a subset of the CSR controller clones > > > > + * which erroneously claim to support it. > > > > + * > > > > + * This quirk must be set before hci_register_dev is called. > > > > + */ > > > > + HCI_QUIRK_BROKEN_READ_PAGE_SCAN_TYPE, > > > > }; > > > > > > > > /* HCI device flags */ > > > > diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c > > > > index dd770ef5ec36..e76012956020 100644 > > > > --- a/net/bluetooth/hci_sync.c > > > > +++ b/net/bluetooth/hci_sync.c > > > > @@ -3696,6 +3696,9 @@ static int hci_read_local_name_sync(struct hci_dev *hdev) > > > > /* Read Voice Setting */ > > > > static int hci_read_voice_setting_sync(struct hci_dev *hdev) > > > > { > > > > + if (test_bit(HCI_QUIRK_BROKEN_READ_VOICE_SETTING, &hdev->quirks)) > > > > + return 0; > > > > > > While at it I'd add the checking for the bit as well, that said > > > perhaps we need to disable SCO link if voice settings cannot be read, > > > is the controller able to create SCO connections? > > > > I did some tests on the version before the regression and with the patches. > > It is possible to create an SCO connection, but the connection is not usable. > > Before the regression there is no sound, and with the patches, dmesg > > is spammed with "Bluetooth: hci0: corrupted SCO packet". > > Then perhaps we should disable SCO support if it doesn't support read > voice settings. I'm having some trouble disabling SCO. We can't add BTUSB_BROKEN_ISOC to the drivers quirks because not all 0a12:0001 are clones with broken READ_VOICE_SETTING. The fake identification is done after the btusb_probe, and we set up the isoc during the btusb_probe. I'm not sure if removing the isoc device file and releasing its interface in the hdev->setup is actually safe. > > > > > > > > return __hci_cmd_sync_status(hdev, HCI_OP_READ_VOICE_SETTING, > > > > 0, NULL, HCI_CMD_TIMEOUT); > > > > } > > > > @@ -4132,6 +4135,9 @@ static int hci_read_page_scan_type_sync(struct hci_dev *hdev) > > > > if (!(hdev->commands[13] & 0x01)) > > > > return 0; > > > > > > > > + if (test_bit(HCI_QUIRK_BROKEN_READ_PAGE_SCAN_TYPE, &hdev->quirks)) > > > > + return 0; > > > > + > > > > return __hci_cmd_sync_status(hdev, HCI_OP_READ_PAGE_SCAN_TYPE, > > > > 0, NULL, HCI_CMD_TIMEOUT); > > > > } > > > > -- > > > > 2.48.1 > > > > > > > > > > > > > -- > > > Luiz Augusto von Dentz > > > > > > -- > > Pedro Nishiyama > > > > -- > Luiz Augusto von Dentz
These fake controllers cannot be initialized because they return a smaller report than expected for HCI_READ_VOICE_SETTING and HCI_READ_PAGE_SCAN_TYPE. This affects fake controllers reusing the 0A12:0001 VID/PID. Fixes: c8992cffbe74 ("Bluetooth: hci_event: Use of a function table to handle Command Complete") Signed-off-by: Pedro Nishiyama <nishiyama.pedro@gmail.com> --- Changes in v3: - Correct the fixes commit. Changes in v2: - Separate the driver changes from the quirks changes. --- Pedro Nishiyama (2): Bluetooth: Add new quirks for fake Bluetooth dongles Bluetooth: btusb: Fix regression in the initialization of fake Bluetooth controllers drivers/bluetooth/btusb.c | 2 ++ include/net/bluetooth/hci.h | 16 ++++++++++++++++ net/bluetooth/hci_sync.c | 6 ++++++ 3 files changed, 24 insertions(+)