Hi Ismael,
> Turns out Hans de Goede completed the work I started last year trying to
> improve Chinese-clone detection of CSR controller chips. Quirk after quirk
> these Bluetooth dongles are more usable now.
>
> Even after a few BlueZ regressions; these clones are so fickle that some
> days they stop working altogether. Except on Windows, they work fine.
>
>
> But this force-suspend initialization quirk seems to mostly do the trick,
> after a lot of testing Bluetooth now seems to work *all* the time.
>
> The only problem is that the solution ended up being masked under a very
> stringent check; when there are probably hundreds of fake dongle
> models out there that benefit from a good reset. Make it so.
>
>
> Fixes: 81cac64ba258a ("Bluetooth: Deal with USB devices that are faking CSR vendor")
> Fixes: cde1a8a992875 ("Bluetooth: btusb: Fix and detect most of the Chinese Bluetooth controllers")
> Fixes: d74e0ae7e0303 ("Bluetooth: btusb: Fix detection of some fake CSR controllers with a bcdDevice val of 0x0134")
> Fixes: 0671c0662383e ("Bluetooth: btusb: Add workaround for remote-wakeup issues with Barrot 8041a02 fake CSR controllers")
>
> Cc: [email protected]
> Cc: Hans de Goede <[email protected]>
> Tested-by: Ismael Ferreras Morezuelas <[email protected]>
> Signed-off-by: Ismael Ferreras Morezuelas <[email protected]>
> ---
>
> I've changed the warning line to make it easy to grep and detect if this updated
> workaround is part of the driver. Should make it much more obvious to users in
> case their dongle doesn't work for other reasons. There's a clear then-now.
>
> Easy to narrow other future issues down. Let me know what you think.
>
> drivers/bluetooth/btusb.c | 61 +++++++++++++++++++++------------------
> 1 file changed, 33 insertions(+), 28 deletions(-)
patch has been applied to bluetooth-next tree.
Regards
Marcel
On 29/07/2021 21:55, Marcel Holtmann wrote:
> Hi Ismael,
>
> patch has been applied to bluetooth-next tree.
>
> Regards
>
> Marcel
>
Thanks a lot, Marcel! Upstreaming these changes has been very fun,
looking forward to improve the compatibility more in the future.
Every bit helps. The next step will probably entail
adding a HCI_FLT_CLEAR_ALL quirk in there. :)