From: Jerome Brunet Subject: Re: [PATCH 1/2] hwrng: meson: expose RNG0 clock via DT Date: Mon, 20 Feb 2017 11:07:23 +0100 Message-ID: <1487585243.3734.3.camel@baylibre.com> References: <00b2a4cc-686e-0cc0-43e8-552b70a1e868@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 8bit Cc: linux-amlogic@lists.infradead.org, linux-crypto@vger.kernel.org, Neil Armstrong To: Heiner Kallweit , Kevin Hilman , Herbert Xu Return-path: Received: from mail-wm0-f52.google.com ([74.125.82.52]:38407 "EHLO mail-wm0-f52.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751232AbdBTKH0 (ORCPT ); Mon, 20 Feb 2017 05:07:26 -0500 Received: by mail-wm0-f52.google.com with SMTP id r141so48292197wmg.1 for ; Mon, 20 Feb 2017 02:07:26 -0800 (PST) In-Reply-To: Sender: linux-crypto-owner@vger.kernel.org List-ID: On Sun, 2017-02-19 at 21:33 +0100, Heiner Kallweit wrote: > Expose the RNG0 clock via DT. > > Signed-off-by: Heiner Kallweit > --- >  arch/arm64/boot/dts/amlogic/meson-gx.dtsi   | 2 +- >  arch/arm64/boot/dts/amlogic/meson-gxbb.dtsi | 5 +++++ >  include/dt-bindings/clock/gxbb-clkc.h       | 1 + >  3 files changed, 7 insertions(+), 1 deletion(-) > > diff --git a/arch/arm64/boot/dts/amlogic/meson-gx.dtsi > b/arch/arm64/boot/dts/amlogic/meson-gx.dtsi > index 5d995f77..620495a4 100644 > --- a/arch/arm64/boot/dts/amlogic/meson-gx.dtsi > +++ b/arch/arm64/boot/dts/amlogic/meson-gx.dtsi > @@ -380,7 +380,7 @@ >   #size-cells = <2>; >   ranges = <0x0 0x0 0x0 0xc8834000 0x0 > 0x2000>; >   > - rng { > + hwrng: rng { >   compatible = "amlogic,meson-rng"; >   reg = <0x0 0x0 0x0 0x4>; >   }; > diff --git a/arch/arm64/boot/dts/amlogic/meson-gxbb.dtsi > b/arch/arm64/boot/dts/amlogic/meson-gxbb.dtsi > index 04b3324b..a375cb21 100644 > --- a/arch/arm64/boot/dts/amlogic/meson-gxbb.dtsi > +++ b/arch/arm64/boot/dts/amlogic/meson-gxbb.dtsi > @@ -524,3 +524,8 @@ >  &vpu { >   compatible = "amlogic,meson-gxbb-vpu", "amlogic,meson-gx- > vpu"; >  }; > + > +&hwrng { > + clocks = <&clkc CLKID_RNG0>; > + clock-names = "core"; > +}; > diff --git a/include/dt-bindings/clock/gxbb-clkc.h b/include/dt- > bindings/clock/gxbb-clkc.h > index 692846c7..473676b1 100644 > --- a/include/dt-bindings/clock/gxbb-clkc.h > +++ b/include/dt-bindings/clock/gxbb-clkc.h > @@ -15,6 +15,7 @@ >  #define CLKID_SPI 34 >  #define CLKID_I2C 22 >  #define CLKID_SAR_ADC 23 > +#define CLKID_RNG0 25 Shouldn't you comment out the corresponding "define" in drivers/clk/meson/gxbb.h ? I don't know if it is rule, but we usually "expose" the clock (moving the define from drivers/clk/meson/gxbb.h to include/dt- bindings/clock/gxbb-clkc.h) in a dedicated patch, with the subject starting with "clk: gxbb: expose foobar clock". It helps the maintainers of the subsystem to give their ack. For this to happen, please remember the patch should be sent to them as well. Here you are missing the clock and dt guys: * Stephen Boyd * Michael Turquette * linux-clk@vger.kernel.org * devicetree@vger.kernel.org scripts/get_maintainer.pl can help you finding all people your patch should be sent to. The remaining part of this patch involves DT, so the subject should start with "ARM64: dts: meson-gx:" You also need to add devicetree@vger.kernel.org in CC. Apart from these minor remarks, the patchset looks good to me. Thx ! >  #define CLKID_ETH 36 >  #define CLKID_USB0 50 >  #define CLKID_USB1 51