Hi Caesar,
Am Dienstag, 11. Juli 2017, 16:04:38 CEST schrieb Caesar Wang:
> Heiko,
>
> 在 2017年07月10日 20:12, Heiko Stuebner 写道:
> > The binding specifies the actual implementations only (mali-t760
> > for example) but not the arm,mali-midgard used in some vendor kernels.
> > So drop that compatible property from the rk3288 where it had slipped in.
> >
> > Signed-off-by: Heiko Stuebner <heiko@xxxxxxxxx>
> > ---
> > arch/arm/boot/dts/rk3288.dtsi | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/arch/arm/boot/dts/rk3288.dtsi b/arch/arm/boot/dts/rk3288.dtsi
> > index cb1d415d1201..e99d7ff4c939 100644
> > --- a/arch/arm/boot/dts/rk3288.dtsi
> > +++ b/arch/arm/boot/dts/rk3288.dtsi
> > @@ -1135,7 +1135,7 @@
> > };
> >
> > gpu: mali@ffa30000 {
> > - compatible = "rockchip,rk3288-mali", "arm,mali-t760", "arm,mali-midgard";
> > + compatible = "rockchip,rk3288-mali", "arm,mali-t760";
>
> If we remove the "arm,mali-midgard", the mali driver maybe need update
> first.
>
> From the lastest MALI driver on ARM supply, the compatible only two
> property.
> mali_kbase_core_linux.c
> 4163: { .compatible = "arm,malit6xx" },
> 4164: { .compatible = "arm,mali-midgard" },
>
> That's odd with the two compatible, the driver should support the more.
Correct, the driver should be updated anyway. In general the official
dt-binding (the one in the linux kernel) should be followed, not the random
out-of-tree one distributed by vendor drivers :-) .
If you look for example at utgard (mali400 for example), the sane binding
used in the kernel, differs a lot from the binding used by the vendor-driver
as well (io-areas, irq-names etc).
In any case, as you use the mali vendor-driver with you vendor-tree,
it should be possible to do both in sync (dt-change from mainline; driver
change to use better compatibles).
Heiko
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
