From mboxrd@z Thu Jan  1 00:00:00 1970
From: sean.wang@mediatek.com (Sean Wang)
Date: Sat, 12 May 2018 13:28:26 +0800
Subject: [PATCH v2 03/12] arm: dts: mt7623: fix invalid memory node
 being generated
In-Reply-To: <4dcb0435-1f98-0753-0feb-d17ad02077ae@gmail.com>
References: <cover.1523434757.git.sean.wang@mediatek.com>
 <d887b72e1839c7e476179e384cfc712216096f20.1523434758.git.sean.wang@mediatek.com>
 <4dcb0435-1f98-0753-0feb-d17ad02077ae@gmail.com>
Message-ID: <1526102906.18877.42.camel@mtkswgap22>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

On Fri, 2018-05-11 at 17:03 +0200, Matthias Brugger wrote:
> 
> On 04/11/2018 10:53 AM, sean.wang at mediatek.com wrote:
> > From: Sean Wang <sean.wang@mediatek.com>
> > 
> > Below two wrong nodes in existing DTS files would cause a fail boot since
> > in fact the address 0 is not the correct place the memory device locates
> > at.
> > 
> > memory {
> >         device_type = "memory";
> >         reg = <0x0 0x0 0x0 0x0>;
> > };
> > 
> > memory at 80000000 {
> >         reg = <0x0 0x80000000 0x0 0x40000000>;
> > };
> > 
> > In order to avoid having a memory node starting at address 0, we can't
> > include file skeleton64.dtsi and instead need to explicitly manually
> > define a few of properties the DTS relies on such as #address-cells
> > and #size-cells in root node and device_type in the node memory at 80000000.
> > 
> > Cc: stable at vger.kernel.org
> > Fixes: 31ac0d69a1d4 ("ARM: dts: mediatek: add MT7623 basic support")
> > Signed-off-by: Sean Wang <sean.wang@mediatek.com>
> > Cc: Rob Herring <robh+dt@kernel.org>
> > ---
> >  arch/arm/boot/dts/mt7623.dtsi                 | 3 ++-
> >  arch/arm/boot/dts/mt7623n-bananapi-bpi-r2.dts | 1 +
> >  arch/arm/boot/dts/mt7623n-rfb.dtsi            | 1 +
> >  3 files changed, 4 insertions(+), 1 deletion(-)
> > 
> 
> merged. We would need this at least for mt2701 as well, correct?
> Would you mind to provide a patch.
> 
> Regards,
> Matthias
> 

Thanks! I totally think the same problem could happen on mt2701, so I'm 
happy to come up with a patch for that.

	Sean

> > diff --git a/arch/arm/boot/dts/mt7623.dtsi b/arch/arm/boot/dts/mt7623.dtsi
> > index fec4715..406a9f3 100644
> > --- a/arch/arm/boot/dts/mt7623.dtsi
> > +++ b/arch/arm/boot/dts/mt7623.dtsi
> > @@ -15,11 +15,12 @@
> >  #include <dt-bindings/phy/phy.h>
> >  #include <dt-bindings/reset/mt2701-resets.h>
> >  #include <dt-bindings/thermal/thermal.h>
> > -#include "skeleton64.dtsi"
> >  
> >  / {
> >  	compatible = "mediatek,mt7623";
> >  	interrupt-parent = <&sysirq>;
> > +	#address-cells = <2>;
> > +	#size-cells = <2>;
> >  
> >  	cpu_opp_table: opp-table {
> >  		compatible = "operating-points-v2";
> > diff --git a/arch/arm/boot/dts/mt7623n-bananapi-bpi-r2.dts b/arch/arm/boot/dts/mt7623n-bananapi-bpi-r2.dts
> > index bbf56f8..5938e4c 100644
> > --- a/arch/arm/boot/dts/mt7623n-bananapi-bpi-r2.dts
> > +++ b/arch/arm/boot/dts/mt7623n-bananapi-bpi-r2.dts
> > @@ -109,6 +109,7 @@
> >  	};
> >  
> >  	memory at 80000000 {
> > +		device_type = "memory";
> >  		reg = <0 0x80000000 0 0x40000000>;
> >  	};
> >  };
> > diff --git a/arch/arm/boot/dts/mt7623n-rfb.dtsi b/arch/arm/boot/dts/mt7623n-rfb.dtsi
> > index a199ae7..343e8ef 100644
> > --- a/arch/arm/boot/dts/mt7623n-rfb.dtsi
> > +++ b/arch/arm/boot/dts/mt7623n-rfb.dtsi
> > @@ -40,6 +40,7 @@
> >  	};
> >  
> >  	memory at 80000000 {
> > +		device_type = "memory";
> >  		reg = <0 0x80000000 0 0x40000000>;
> >  	};
> >  
> >