| Message ID | 1437691862-21312-2-git-send-email-moritz.fischer@ettus.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On 07/24/2015 12:51 AM, Moritz Fischer wrote: > Signed-off-by: Moritz Fischer <moritz.fischer@ettus.com> > --- > Documentation/devicetree/bindings/reset/zynq-reset-pl.txt | 13 +++++++++++++ > 1 file changed, 13 insertions(+) > create mode 100644 Documentation/devicetree/bindings/reset/zynq-reset-pl.txt > > diff --git a/Documentation/devicetree/bindings/reset/zynq-reset-pl.txt b/Documentation/devicetree/bindings/reset/zynq-reset-pl.txt > new file mode 100644 > index 0000000..ac4499e > --- /dev/null > +++ b/Documentation/devicetree/bindings/reset/zynq-reset-pl.txt > @@ -0,0 +1,13 @@ > +Xilinx Zynq PL Reset Manager I think there is no reason to be just PL specific. > + > +Required properties: > +- compatible: "xlnx,zynq-reset-pl" > +- syscon <&slcr>; > +- #reset-cells: 1 > + > +Example: > + rstc: rstc@240 { @200 then here > + #reset-cells = <1>; > + compatible = "xlnx,zynq-reset-pl"; Compatible should be first below node name. You should add also reg property 0x200 0x50 > + syscon = <&slcr>; > + }; > Thanks, Michal
Am Donnerstag, den 23.07.2015, 15:51 -0700 schrieb Moritz Fischer: > Signed-off-by: Moritz Fischer <moritz.fischer@ettus.com> > --- > Documentation/devicetree/bindings/reset/zynq-reset-pl.txt | 13 +++++++++++++ > 1 file changed, 13 insertions(+) > create mode 100644 Documentation/devicetree/bindings/reset/zynq-reset-pl.txt > > diff --git a/Documentation/devicetree/bindings/reset/zynq-reset-pl.txt b/Documentation/devicetree/bindings/reset/zynq-reset-pl.txt > new file mode 100644 > index 0000000..ac4499e > --- /dev/null > +++ b/Documentation/devicetree/bindings/reset/zynq-reset-pl.txt > @@ -0,0 +1,13 @@ > +Xilinx Zynq PL Reset Manager > + > +Required properties: > +- compatible: "xlnx,zynq-reset-pl" > +- syscon <&slcr>; > +- #reset-cells: 1 > + > +Example: > + rstc: rstc@240 { > + #reset-cells = <1>; > + compatible = "xlnx,zynq-reset-pl"; > + syscon = <&slcr>; > + }; Instead of the syscon property, why not specify that the rstc node must be a child of the slcr node? regards Philipp
On 07/24/2015 09:25 AM, Philipp Zabel wrote: > Am Donnerstag, den 23.07.2015, 15:51 -0700 schrieb Moritz Fischer: >> Signed-off-by: Moritz Fischer <moritz.fischer@ettus.com> >> --- >> Documentation/devicetree/bindings/reset/zynq-reset-pl.txt | 13 +++++++++++++ >> 1 file changed, 13 insertions(+) >> create mode 100644 Documentation/devicetree/bindings/reset/zynq-reset-pl.txt >> >> diff --git a/Documentation/devicetree/bindings/reset/zynq-reset-pl.txt b/Documentation/devicetree/bindings/reset/zynq-reset-pl.txt >> new file mode 100644 >> index 0000000..ac4499e >> --- /dev/null >> +++ b/Documentation/devicetree/bindings/reset/zynq-reset-pl.txt >> @@ -0,0 +1,13 @@ >> +Xilinx Zynq PL Reset Manager >> + >> +Required properties: >> +- compatible: "xlnx,zynq-reset-pl" >> +- syscon <&slcr>; >> +- #reset-cells: 1 >> + >> +Example: >> + rstc: rstc@240 { >> + #reset-cells = <1>; >> + compatible = "xlnx,zynq-reset-pl"; >> + syscon = <&slcr>; >> + }; > > Instead of the syscon property, why not specify that the rstc node must > be a child of the slcr node? FYI: It is already a child node of SLCR if you look at location in DTS. Thanks, Michal
On Fri, 2015-07-24 at 06:40AM +0200, Michal Simek wrote: > On 07/24/2015 12:51 AM, Moritz Fischer wrote: > > Signed-off-by: Moritz Fischer <moritz.fischer@ettus.com> > > --- > > Documentation/devicetree/bindings/reset/zynq-reset-pl.txt | 13 +++++++++++++ > > 1 file changed, 13 insertions(+) > > create mode 100644 Documentation/devicetree/bindings/reset/zynq-reset-pl.txt > > > > diff --git a/Documentation/devicetree/bindings/reset/zynq-reset-pl.txt b/Documentation/devicetree/bindings/reset/zynq-reset-pl.txt > > new file mode 100644 > > index 0000000..ac4499e > > --- /dev/null > > +++ b/Documentation/devicetree/bindings/reset/zynq-reset-pl.txt > > @@ -0,0 +1,13 @@ > > +Xilinx Zynq PL Reset Manager > > I think there is no reason to be just PL specific. That was my first thought too. Why not model all the resets in the SLCR? Sören
Michal, Sören, On Fri, Jul 24, 2015 at 7:50 AM, Sören Brinkmann <soren.brinkmann@xilinx.com> wrote: > On Fri, 2015-07-24 at 06:40AM +0200, Michal Simek wrote: >> On 07/24/2015 12:51 AM, Moritz Fischer wrote: >> > Signed-off-by: Moritz Fischer <moritz.fischer@ettus.com> >> > --- >> > Documentation/devicetree/bindings/reset/zynq-reset-pl.txt | 13 +++++++++++++ >> > 1 file changed, 13 insertions(+) >> > create mode 100644 Documentation/devicetree/bindings/reset/zynq-reset-pl.txt >> > >> > diff --git a/Documentation/devicetree/bindings/reset/zynq-reset-pl.txt b/Documentation/devicetree/bindings/reset/zynq-reset-pl.txt >> > new file mode 100644 >> > index 0000000..ac4499e >> > --- /dev/null >> > +++ b/Documentation/devicetree/bindings/reset/zynq-reset-pl.txt >> > @@ -0,0 +1,13 @@ >> > +Xilinx Zynq PL Reset Manager >> >> I think there is no reason to be just PL specific. > > That was my first thought too. Why not model all the resets in the SLCR? I only needed the ones for the PL for my fpga-mgr work and reading the TRM had a hard time to decide which ones make sense, and which ones don't make sense to expose to Linux. I'll look into reworking it to support all the resets. > Sören Moritz
diff --git a/Documentation/devicetree/bindings/reset/zynq-reset-pl.txt b/Documentation/devicetree/bindings/reset/zynq-reset-pl.txt new file mode 100644 index 0000000..ac4499e --- /dev/null +++ b/Documentation/devicetree/bindings/reset/zynq-reset-pl.txt @@ -0,0 +1,13 @@ +Xilinx Zynq PL Reset Manager + +Required properties: +- compatible: "xlnx,zynq-reset-pl" +- syscon <&slcr>; +- #reset-cells: 1 + +Example: + rstc: rstc@240 { + #reset-cells = <1>; + compatible = "xlnx,zynq-reset-pl"; + syscon = <&slcr>; + };
Signed-off-by: Moritz Fischer <moritz.fischer@ettus.com> --- Documentation/devicetree/bindings/reset/zynq-reset-pl.txt | 13 +++++++++++++ 1 file changed, 13 insertions(+) create mode 100644 Documentation/devicetree/bindings/reset/zynq-reset-pl.txt