On Mon, Oct 12, 2020 at 09:54:25AM +0200, Neil Armstrong wrote: > Hi, > > On 11/10/2020 00:41, Serge Semin wrote: > > An empty snps,quirk-frame-length-adjustment won't cause any change > > performed by the driver. Moreover the DT schema validation will fail, > > since it expects the property being assigned with some value. So just > > discard the property declaration then from the example. > > > > Signed-off-by: Serge Semin <Sergey.Semin@xxxxxxxxxxxxxxxxxxxx> > > > > --- > > > > Note the same problem is in the DT source file > > arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi . > > --- > > .../devicetree/bindings/usb/amlogic,meson-g12a-usb-ctrl.yaml | 1 - > > 1 file changed, 1 deletion(-) > > > > diff --git a/Documentation/devicetree/bindings/usb/amlogic,meson-g12a-usb-ctrl.yaml b/Documentation/devicetree/bindings/usb/amlogic,meson-g12a-usb-ctrl.yaml > > index 5b04a7dfa018..88184d7e26cc 100644 > > --- a/Documentation/devicetree/bindings/usb/amlogic,meson-g12a-usb-ctrl.yaml > > +++ b/Documentation/devicetree/bindings/usb/amlogic,meson-g12a-usb-ctrl.yaml > > @@ -209,6 +209,5 @@ examples: > > interrupts = <30>; > > dr_mode = "host"; > > snps,dis_u2_susphy_quirk; > > - snps,quirk-frame-length-adjustment; > > }; > > }; > > > > Thanks for reporting this, actually the fladj must be 0x20 on this hw, > but we do set this on the PHY side, so we can let the dwc3 side 0 here. I can convert this patch to initializing the "snps,quirk-frame-length-adjustment" property with 0x20 value instead. Since most likely I'll have to send a v2/v3/etc of this patchset, that modification won't be too much work to do. What do you think? Anyway please note, that I've fixed the improper property usage in the DT schema example only. "snps,quirk-frame-length-adjustment" defined as boolean still persists in the DTS file: arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi . So if you ever try to validate that dts file with "make dtbs_check" scenario, it will fail. -Sergey > > Acked-by: Neil Armstrong <narmstrong@xxxxxxxxxxxx> > > Neil >
