| Message ID | 1601712179-13540-1-git-send-email-kiran.k@intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [v5,1/2] Bluetooth: btusb: use usb vid/pid for initializing hal callbacks | expand |
Hi Kiran, > For Intel controllers, use vid/pid for initalizing hardware abstraction > layer callbacks to avoid defining new quirk flags for new products. > > Signed-off-by: Kiran K <kiran.k@intel.com> > --- > Changes in v5: > * Use usb vid/pid combination to identify controller type istead of using flags > > Changes in previous versions: > None. This is a new patch created part of v5 > > drivers/bluetooth/btusb.c | 79 ++++++++++++++++++++++++++++------------------- > 1 file changed, 48 insertions(+), 31 deletions(-) > > diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c > index 1005b6e..e2cd78d 100644 > --- a/drivers/bluetooth/btusb.c > +++ b/drivers/bluetooth/btusb.c > @@ -3969,13 +3969,21 @@ static int btusb_probe(struct usb_interface *intf, > init_usb_anchor(&data->ctrl_anchor); > spin_lock_init(&data->rxlock); > > - if (id->driver_info & BTUSB_INTEL_NEW) { > - data->recv_event = btusb_recv_event_intel; > - data->recv_bulk = btusb_recv_bulk_intel; > - set_bit(BTUSB_BOOTLOADER, &data->flags); > - } else { > - data->recv_event = hci_recv_frame; > - data->recv_bulk = btusb_recv_bulk; > + data->recv_event = hci_recv_frame; > + data->recv_bulk = btusb_recv_bulk; > + > + if (id->idVendor == 0x8087) { > + switch (id->idProduct) { > + case 0x0025: > + case 0x0026: > + case 0x0029: > + case 0x0a2b: > + case 0x0aaa: > + data->recv_event = btusb_recv_event_intel; > + data->recv_bulk = btusb_recv_bulk_intel; > + set_bit(BTUSB_BOOTLOADER, &data->flags); > + break; > + } > } please don’t do this. This is exactly what I didn’t want. Listing PID in the device table and then also again in the probe callback is bound to be error prone. Regards Marcel
Hi Marcel, > -----Original Message----- > From: Marcel Holtmann <marcel@holtmann.org> > Sent: Saturday, October 3, 2020 1:57 PM > To: Kiran K <kiraank@gmail.com> > Cc: linux-bluetooth <linux-bluetooth@vger.kernel.org>; Narasimman, Sathish > <sathish.narasimman@intel.com>; Tumkur Narayan, Chethan > <chethan.tumkur.narayan@intel.com>; Srivatsa, Ravishankar > <ravishankar.srivatsa@intel.com>; K, Kiran <kiran.k@intel.com> > Subject: Re: [PATCH v5 1/2] Bluetooth: btusb: use usb vid/pid for initializing > hal callbacks > > Hi Kiran, > > > For Intel controllers, use vid/pid for initalizing hardware > > abstraction layer callbacks to avoid defining new quirk flags for new > products. > > > > Signed-off-by: Kiran K <kiran.k@intel.com> > > --- > > Changes in v5: > > * Use usb vid/pid combination to identify controller type istead of > > using flags > > > > Changes in previous versions: > > None. This is a new patch created part of v5 > > > > drivers/bluetooth/btusb.c | 79 > > ++++++++++++++++++++++++++++------------------- > > 1 file changed, 48 insertions(+), 31 deletions(-) > > > > diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c > > index 1005b6e..e2cd78d 100644 > > --- a/drivers/bluetooth/btusb.c > > +++ b/drivers/bluetooth/btusb.c > > @@ -3969,13 +3969,21 @@ static int btusb_probe(struct usb_interface > *intf, > > init_usb_anchor(&data->ctrl_anchor); > > spin_lock_init(&data->rxlock); > > > > - if (id->driver_info & BTUSB_INTEL_NEW) { > > - data->recv_event = btusb_recv_event_intel; > > - data->recv_bulk = btusb_recv_bulk_intel; > > - set_bit(BTUSB_BOOTLOADER, &data->flags); > > - } else { > > - data->recv_event = hci_recv_frame; > > - data->recv_bulk = btusb_recv_bulk; > > + data->recv_event = hci_recv_frame; > > + data->recv_bulk = btusb_recv_bulk; > > + > > + if (id->idVendor == 0x8087) { > > + switch (id->idProduct) { > > + case 0x0025: > > + case 0x0026: > > + case 0x0029: > > + case 0x0a2b: > > + case 0x0aaa: > > + data->recv_event = btusb_recv_event_intel; > > + data->recv_bulk = btusb_recv_bulk_intel; > > + set_bit(BTUSB_BOOTLOADER, &data->flags); > > + break; > > + } > > } > > please don’t do this. This is exactly what I didn’t want. Listing PID in the > device table and then also again in the probe callback is bound to be error > prone. > Got your point. I will send the fix with BTUSB_INTEL_NEWGEN as discussed earlier. Thanks. > Regards > > Marcel Regards, Kiran
diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index 1005b6e..e2cd78d 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -3969,13 +3969,21 @@ static int btusb_probe(struct usb_interface *intf, init_usb_anchor(&data->ctrl_anchor); spin_lock_init(&data->rxlock); - if (id->driver_info & BTUSB_INTEL_NEW) { - data->recv_event = btusb_recv_event_intel; - data->recv_bulk = btusb_recv_bulk_intel; - set_bit(BTUSB_BOOTLOADER, &data->flags); - } else { - data->recv_event = hci_recv_frame; - data->recv_bulk = btusb_recv_bulk; + data->recv_event = hci_recv_frame; + data->recv_bulk = btusb_recv_bulk; + + if (id->idVendor == 0x8087) { + switch (id->idProduct) { + case 0x0025: + case 0x0026: + case 0x0029: + case 0x0a2b: + case 0x0aaa: + data->recv_event = btusb_recv_event_intel; + data->recv_bulk = btusb_recv_bulk_intel; + set_bit(BTUSB_BOOTLOADER, &data->flags); + break; + } } hdev = hci_alloc_dev(); @@ -4052,30 +4060,39 @@ static int btusb_probe(struct usb_interface *intf, data->diag = usb_ifnum_to_if(data->udev, ifnum_base + 2); } - if (id->driver_info & BTUSB_INTEL) { - hdev->manufacturer = 2; - hdev->setup = btusb_setup_intel; - hdev->shutdown = btusb_shutdown_intel; - hdev->set_diag = btintel_set_diag_mfg; - hdev->set_bdaddr = btintel_set_bdaddr; - hdev->cmd_timeout = btusb_intel_cmd_timeout; - set_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks); - set_bit(HCI_QUIRK_SIMULTANEOUS_DISCOVERY, &hdev->quirks); - set_bit(HCI_QUIRK_NON_PERSISTENT_DIAG, &hdev->quirks); - } - - if (id->driver_info & BTUSB_INTEL_NEW) { - hdev->manufacturer = 2; - hdev->send = btusb_send_frame_intel; - hdev->setup = btusb_setup_intel_new; - hdev->shutdown = btusb_shutdown_intel_new; - hdev->hw_error = btintel_hw_error; - hdev->set_diag = btintel_set_diag; - hdev->set_bdaddr = btintel_set_bdaddr; - hdev->cmd_timeout = btusb_intel_cmd_timeout; - set_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks); - set_bit(HCI_QUIRK_SIMULTANEOUS_DISCOVERY, &hdev->quirks); - set_bit(HCI_QUIRK_NON_PERSISTENT_DIAG, &hdev->quirks); + if (id->idVendor == 0x8087) { + switch (id->idProduct) { + case 0x07dc: + case 0x0a2a: + case 0x0aa7: + hdev->manufacturer = 2; + hdev->setup = btusb_setup_intel; + hdev->shutdown = btusb_shutdown_intel; + hdev->set_diag = btintel_set_diag_mfg; + hdev->set_bdaddr = btintel_set_bdaddr; + hdev->cmd_timeout = btusb_intel_cmd_timeout; + set_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks); + set_bit(HCI_QUIRK_SIMULTANEOUS_DISCOVERY, &hdev->quirks); + set_bit(HCI_QUIRK_NON_PERSISTENT_DIAG, &hdev->quirks); + break; + case 0x0025: + case 0x0026: + case 0x0029: + case 0x0a2b: + case 0x0aaa: + hdev->manufacturer = 2; + hdev->send = btusb_send_frame_intel; + hdev->setup = btusb_setup_intel_new; + hdev->shutdown = btusb_shutdown_intel_new; + hdev->hw_error = btintel_hw_error; + hdev->set_diag = btintel_set_diag; + hdev->set_bdaddr = btintel_set_bdaddr; + hdev->cmd_timeout = btusb_intel_cmd_timeout; + set_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks); + set_bit(HCI_QUIRK_SIMULTANEOUS_DISCOVERY, &hdev->quirks); + set_bit(HCI_QUIRK_NON_PERSISTENT_DIAG, &hdev->quirks); + break; + } } if (id->driver_info & BTUSB_MARVELL)
For Intel controllers, use vid/pid for initalizing hardware abstraction layer callbacks to avoid defining new quirk flags for new products. Signed-off-by: Kiran K <kiran.k@intel.com> --- Changes in v5: * Use usb vid/pid combination to identify controller type istead of using flags Changes in previous versions: None. This is a new patch created part of v5 drivers/bluetooth/btusb.c | 79 ++++++++++++++++++++++++++++------------------- 1 file changed, 48 insertions(+), 31 deletions(-)