| Message ID | 20150915004827.GA19862@home.buserror.net (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
> From: linuxppc-release-bounces@linux.freescale.net [mailto:linuxppc- > release-bounces@linux.freescale.net] On Behalf Of Scott Wood > Sent: Tuesday, September 15, 2015 6:18 AM > > These will allow U-Boot to set linux,stdout-path in /chosen based on the > active console device, which will allow devicetree-based earlycon. > > Signed-off-by: Scott Wood <scottwood@freescale.com> > --- > arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi | 5 +++++ > 1 file changed, 5 insertions(+) > > diff --git a/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi > b/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi > index e281ceb..eae454e 100644 > --- a/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi > +++ b/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi > @@ -55,6 +55,11 @@ > #address-cells = <2>; > #size-cells = <2>; > > + aliases { > + serial0 = &serial0; > + serial1 = &serial1; > + }; > + > cpus { > #address-cells = <2>; > #size-cells = <0>; > -- Rebase this to my LS2080A v2 DTS sent some time ago, as LS2085A is now renamed to LS2080A. Regards, Bhupesh
On Tue, 2015-09-15 at 00:57 -0500, Sharma Bhupesh-B45370 wrote: > > From: linuxppc-release-bounces@linux.freescale.net[mailto:linuxppc- > > release-bounces@linux.freescale.net] On Behalf Of Scott Wood > > Sent: Tuesday, September 15, 2015 6:18 AM > > > > These will allow U-Boot to set linux,stdout-path in /chosen based on the > > active console device, which will allow devicetree-based earlycon. > > > > Signed-off-by: Scott Wood <scottwood@freescale.com> > > --- > > arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi | 5 +++++ > > 1 file changed, 5 insertions(+) > > > > diff --git a/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi > > b/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi > > index e281ceb..eae454e 100644 > > --- a/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi > > +++ b/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi > > @@ -55,6 +55,11 @@ > > #address-cells = <2>; > > #size-cells = <2>; > > > > + aliases { > > + serial0 = &serial0; > > + serial1 = &serial1; > > + }; > > + > > cpus { > > #address-cells = <2>; > > #size-cells = <0>; > > -- > > Rebase this to my LS2080A v2 DTS sent some time ago, as LS2085A is now > renamed to LS2080A. Patchwork link? How long ago? When is it going to be merged? -Scott
On Monday 14 September 2015 19:48:27 Scott Wood wrote: > These will allow U-Boot to set linux,stdout-path in /chosen based on > the active console device, which will allow devicetree-based earlycon. > > Signed-off-by: Scott Wood <scottwood@freescale.com> > --- > arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi | 5 +++++ > 1 file changed, 5 insertions(+) > > diff --git a/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi b/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi > index e281ceb..eae454e 100644 > --- a/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi > +++ b/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi > @@ -55,6 +55,11 @@ > #address-cells = <2>; > #size-cells = <2>; > > + aliases { > + serial0 = &serial0; > + serial1 = &serial1; > + }; > + I normally ask people to put the aliases into the board specific dts file, to account for machines that do not have all the uarts, or that number them differently from the SoC. I didn't know about U-Boot setting the stdout-path property, but that's great, as long as U-Boot uses the same numbering of the uarts that the board uses. Do you know if that is normally the case? Arnd
On Tue, 2015-09-15 at 08:43 +0200, Arnd Bergmann wrote: > On Monday 14 September 2015 19:48:27 Scott Wood wrote: > > These will allow U-Boot to set linux,stdout-path in /chosen based on > > the active console device, which will allow devicetree-based earlycon. > > > > Signed-off-by: Scott Wood <scottwood@freescale.com> > > --- > > arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi | 5 +++++ > > 1 file changed, 5 insertions(+) > > > > diff --git a/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi > > b/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi > > index e281ceb..eae454e 100644 > > --- a/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi > > +++ b/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi > > @@ -55,6 +55,11 @@ > > #address-cells = <2>; > > #size-cells = <2>; > > > > + aliases { > > + serial0 = &serial0; > > + serial1 = &serial1; > > + }; > > + > > I normally ask people to put the aliases into the board specific dts file, > to account for machines that do not have all the uarts, or that number them > differently from the SoC. OK, I'll respin on top of https://patchwork.kernel.org/patch/7120211/ > I didn't know about U-Boot setting the stdout-path property, but that's > great, as long as U-Boot uses the same numbering of the uarts that the > board uses. Do you know if that is normally the case? Yes, the numbering of the serial aliases in the dts should match the numbering that U-Boot uses. -Scott
diff --git a/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi b/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi index e281ceb..eae454e 100644 --- a/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi +++ b/arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi @@ -55,6 +55,11 @@ #address-cells = <2>; #size-cells = <2>; + aliases { + serial0 = &serial0; + serial1 = &serial1; + }; + cpus { #address-cells = <2>; #size-cells = <0>;
These will allow U-Boot to set linux,stdout-path in /chosen based on the active console device, which will allow devicetree-based earlycon. Signed-off-by: Scott Wood <scottwood@freescale.com> --- arch/arm64/boot/dts/freescale/fsl-ls2085a.dtsi | 5 +++++ 1 file changed, 5 insertions(+)