Hi Marcel,
On Tue, Oct 12, 2021 at 11:35 PM Marcel Holtmann <marcel@xxxxxxxxxxxx> wrote:
>
> Hi Luiz,
>
> > The msft_opcode shall only be changed while at the setup stage otherwise
> > it can possible cause problems where different opcodes are used while
> > running.
> >
> > Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@xxxxxxxxx>
> > ---
> > v2: Fix typos: s/extention/extension/g
> >
> > include/net/bluetooth/hci_core.h | 6 +++++-
> > 1 file changed, 5 insertions(+), 1 deletion(-)
> >
> > diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
> > index dd8840e70e25..eb5d4ea88c3a 100644
> > --- a/include/net/bluetooth/hci_core.h
> > +++ b/include/net/bluetooth/hci_core.h
> > @@ -1272,11 +1272,15 @@ int hci_recv_diag(struct hci_dev *hdev, struct sk_buff *skb);
> > __printf(2, 3) void hci_set_hw_info(struct hci_dev *hdev, const char *fmt, ...);
> > __printf(2, 3) void hci_set_fw_info(struct hci_dev *hdev, const char *fmt, ...);
> >
> > -static inline void hci_set_msft_opcode(struct hci_dev *hdev, __u16 opcode)
> > +static inline int hci_set_msft_opcode(struct hci_dev *hdev, __u16 opcode)
> > {
> > + if (!hci_dev_test_flag(hdev, HCI_SETUP))
> > + return -EPERM;
> > +
> > #if IS_ENABLED(CONFIG_BT_MSFTEXT)
> > hdev->msft_opcode = opcode;
> > #endif
> > + return 0;
> > }
>
> this is also not going to work since some driver might set it in their probe() routine before calling hci_register_dev.
Shall we allow that though? Perhaps we should be checking for HCI_RUNNING then?
--
Luiz Augusto von Dentz
