Re: [PATCH v3 4/5] Bluetooth: advmon offload MSFT handle controller reset

[Archie Pusaka]

Hi Marcel,

I've sent a new v5 patch to address this issue.

Thanks,
Archie

On Tue, 22 Dec 2020 at 18:03, Marcel Holtmann <marcel@xxxxxxxxxxxx> wrote:
>
> Hi Archie,
>
> >>> When the controller is powered off, the registered advertising monitor
> >>> is removed from the controller. This patch handles the re-registration
> >>> of those monitors when the power is on.
> >>>
> >>> Signed-off-by: Archie Pusaka <apusaka@xxxxxxxxxxxx>
> >>> Reviewed-by: Miao-chen Chou <mcchou@xxxxxxxxxxxx>
> >>> Reviewed-by: Yun-Hao Chung <howardchung@xxxxxxxxxx>
> >>>
> >>> ---
> >>>
> >>> (no changes since v1)
> >>>
> >>> net/bluetooth/msft.c | 79 +++++++++++++++++++++++++++++++++++++++++---
> >>> 1 file changed, 74 insertions(+), 5 deletions(-)
> >>>
> >>> diff --git a/net/bluetooth/msft.c b/net/bluetooth/msft.c
> >>> index f5aa0e3b1b9b..7e33a85c3f1c 100644
> >>> --- a/net/bluetooth/msft.c
> >>> +++ b/net/bluetooth/msft.c
> >>> @@ -82,8 +82,15 @@ struct msft_data {
> >>>      struct list_head handle_map;
> >>>      __u16 pending_add_handle;
> >>>      __u16 pending_remove_handle;
> >>> +
> >>> +     struct {
> >>> +             u8 reregistering:1;
> >>> +     } flags;
> >>> };
> >>
> >> hmmm. Do you have bigger plans with this struct? I would just skip it.
> >>
> > This struct is also used in patch 5/5 to store the "enabled" status of
> > the filter.
> > Suspend/resume would need to enable/disable the filter, but it is not
> > yet implemented in this patch series.
>
> just do it without the nested structs. I think you are overdoing it here.
>
> Regards
>
> Marcel
>