On Tue, Sep 25, 2018 at 11:55:48AM +0200, Arend van Spriel wrote:
> On 9/25/2018 11:48 AM, Stanislaw Gruszka wrote:
> >On Tue, Sep 25, 2018 at 11:07:47AM +0200, Lorenzo Bianconi wrote:
> >>On Sep 25, Felix Fietkau wrote:
> >>>On 2018-09-25 09:54, Lorenzo Bianconi wrote:
> >>>>>
> >>>>>diff --git a/drivers/net/wireless/mediatek/mt76/mt76x0/usb.c b/drivers/net/wireless/mediatek/mt76/mt76x0/usb.c
> >>>>>index 76d607f73758..b7a1069ecd0e 100644
> >>>>>--- a/drivers/net/wireless/mediatek/mt76/mt76x0/usb.c
> >>>>>+++ b/drivers/net/wireless/mediatek/mt76/mt76x0/usb.c
> >>>>>@@ -45,7 +45,8 @@
> >>>>> { USB_DEVICE(0x20f4, 0x806b) }, /* TRENDnet TEW-806UBH */
> >>>>> { USB_DEVICE(0x7392, 0xc711) }, /* Devolo Wifi ac Stick */
> >>>>> { USB_DEVICE(0x0df6, 0x0079) }, /* Sitecom Europe B.V. ac Stick */
> >>>>>- { USB_DEVICE(0x2357, 0x0105) }, /* TP-LINK Archer T1U */
> >>>>>+ { USB_DEVICE(0x2357, 0x0105),
> >>>>>+ .driver_info = 1, }, /* TP-LINK Archer T1U */
> >>>>
> >>>>Hi Stanislaw,
> >>>>
> >>>>what about using 'ieee80211-freq-limit' OF property to limit available wireless
> >>>>band? In this way we can take into account even the opposite case (no 5GHz).
> >>>That doesn't make sense for USB devices, which can be plugged into any
> >>>machine.
> >>
> >>Ack, right.
> >>What about a more general solution like adding an enum { NO_5GHz = 1, NO_2GHz }?
> >>Does it worth to implement it now?
> >
> >I do not see reason for that. Basically band information should be
> >read from EEPROM, I do not expect need for more quirks like this.
>
> Well. Never say never :-p And the phrase "quirks like this" does
> seem to leave the door open for other quirks. Or did you mean
> "quirks using driver_info"?
I mean quirks that mask out band. Anyway if the need for any other
quirk will show up then the things can be modified. Do not see point
for doing it now, when we don't know if new stuff will be used
at all.
Thanks
Stanislaw
