This patch enables Mediatek to support the AOSP extension.
Reviewed-by: Miao-chen Chou <[email protected]>
Signed-off-by: Joseph Hwang <[email protected]>
---
(no changes since v5)
Changes in v5:
- Let the vendor command in aosp_do_open() to determine what
capabilities are supported.
Changes in v4:
- Call hci_set_aosp_capable in the driver.
- This patch is added in this Series-changes 4.
drivers/bluetooth/btusb.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 87b71740fad8..00311ebd7823 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -3876,6 +3876,8 @@ static int btusb_probe(struct usb_interface *intf,
hdev->set_bdaddr = btusb_set_bdaddr_mtk;
set_bit(HCI_QUIRK_NON_PERSISTENT_SETUP, &hdev->quirks);
data->recv_acl = btusb_recv_acl_mtk;
+
+ hci_set_aosp_capable(hdev);
}
if (id->driver_info & BTUSB_SWAVE) {
--
2.33.0.1079.g6e70778dc9-goog
[email protected]
Some of Mediatek's controllers/firmwares do NOT support AOSP extensions.
Please refer to this Gerrit patch review link for chrome os
(https://chromium-review.googlesource.com/c/chromiumos/third_party/kernel/+/3054233/8/drivers/bluetooth/btusb.c)
which was reviewed by Mark-YW Chen, Mediatek's engineer.
As an example, MediaTek MT7615E does not support the AOSP extensions.
That was why I used a new BTUSB_QUALITY_REPORT flag in the changes v4
and earlier ones.
Please let me know how you would like to proceed with this patch.
Thanks!
Joseph
On Mon, Oct 25, 2021 at 9:26 PM Marcel Holtmann <[email protected]> wrote:
>
> Hi Joseph,
>
> > This patch enables Mediatek to support the AOSP extension.
> >
> > Reviewed-by: Miao-chen Chou <[email protected]>
> > Signed-off-by: Joseph Hwang <[email protected]>
> >
> > ---
> >
> > (no changes since v5)
> >
> > Changes in v5:
> > - Let the vendor command in aosp_do_open() to determine what
> > capabilities are supported.
> >
> > Changes in v4:
> > - Call hci_set_aosp_capable in the driver.
> > - This patch is added in this Series-changes 4.
> >
> > drivers/bluetooth/btusb.c | 2 ++
> > 1 file changed, 2 insertions(+)
> >
> > diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
> > index 87b71740fad8..00311ebd7823 100644
> > --- a/drivers/bluetooth/btusb.c
> > +++ b/drivers/bluetooth/btusb.c
> > @@ -3876,6 +3876,8 @@ static int btusb_probe(struct usb_interface *intf,
> > hdev->set_bdaddr = btusb_set_bdaddr_mtk;
> > set_bit(HCI_QUIRK_NON_PERSISTENT_SETUP, &hdev->quirks);
> > data->recv_acl = btusb_recv_acl_mtk;
> > +
> > + hci_set_aosp_capable(hdev);
>
> can we get an ACK from Mediatek people that all their firmwares support the AOSP extensions.
>
> Regards
>
> Marcel
>
--
Joseph Shyh-In Hwang
Email: [email protected]
Hi Joseph,
> Some of Mediatek's controllers/firmwares do NOT support AOSP extensions.
>
> Please refer to this Gerrit patch review link for chrome os
> (https://chromium-review.googlesource.com/c/chromiumos/third_party/kernel/+/3054233/8/drivers/bluetooth/btusb.c)
> which was reviewed by Mark-YW Chen, Mediatek's engineer.
>
> As an example, MediaTek MT7615E does not support the AOSP extensions.
>
> That was why I used a new BTUSB_QUALITY_REPORT flag in the changes v4
> and earlier ones.
>
> Please let me know how you would like to proceed with this patch.
I want that inside the Mediatek specific hdev->setup() procedure, they decide if AOSP extensions are supported or not. The BTUSB flag business is getting out of hand. I really don’t want that in btusb.c anymore. I moved Intel over to do that in btintel.c and also Realtek is doing it in btrtl.c for their hardware with the MSFT extensions.
We are not doing try-and-error of HCI commands. Just keep that in mind since I said that multiple times now.
Regards
Marcel
Hi Marcel:
I see your point. Thank you for letting us know the right way to go.
I will discuss with Mediatek and request them to implement their
specific setup() for the purpose. And then they will submit a new
patch to enable AOSP extensions for their hardware.
Thanks and regards,
Joseph
On Fri, Oct 29, 2021 at 4:21 PM Marcel Holtmann <[email protected]> wrote:
>
> Hi Joseph,
>
> > Some of Mediatek's controllers/firmwares do NOT support AOSP extensions.
> >
> > Please refer to this Gerrit patch review link for chrome os
> > (https://chromium-review.googlesource.com/c/chromiumos/third_party/kernel/+/3054233/8/drivers/bluetooth/btusb.c)
> > which was reviewed by Mark-YW Chen, Mediatek's engineer.
> >
> > As an example, MediaTek MT7615E does not support the AOSP extensions.
> >
> > That was why I used a new BTUSB_QUALITY_REPORT flag in the changes v4
> > and earlier ones.
> >
> > Please let me know how you would like to proceed with this patch.
>
> I want that inside the Mediatek specific hdev->setup() procedure, they decide if AOSP extensions are supported or not. The BTUSB flag business is getting out of hand. I really don’t want that in btusb.c anymore. I moved Intel over to do that in btintel.c and also Realtek is doing it in btrtl.c for their hardware with the MSFT extensions.
>
> We are not doing try-and-error of HCI commands. Just keep that in mind since I said that multiple times now.
>
> Regards
>
> Marcel
>
--
Joseph Shyh-In Hwang
Email: [email protected]