| Message ID | 1445035350-3569-3-git-send-email-moritz.fischer@ettus.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Fri, Oct 16, 2015 at 03:42:29PM -0700, Moritz Fischer wrote: > Added addtional nodes required for FPGA Manager operation > of the Xilinx Zynq Devc configuration interface. > > Reviewed-by: Sören Brinkmann <soren.brinkmann@xilinx.com> > Signed-off-by: Moritz Fischer <moritz.fischer@ettus.com> > --- > > v2: No changes > > --- > arch/arm/boot/dts/zynq-7000.dtsi | 5 +++++ > 1 file changed, 5 insertions(+) > > diff --git a/arch/arm/boot/dts/zynq-7000.dtsi b/arch/arm/boot/dts/zynq-7000.dtsi > index dc0457e..1a5220e 100644 > --- a/arch/arm/boot/dts/zynq-7000.dtsi > +++ b/arch/arm/boot/dts/zynq-7000.dtsi > @@ -294,6 +294,11 @@ > devcfg: devcfg@f8007000 { > compatible = "xlnx,zynq-devcfg-1.0"; > reg = <0xf8007000 0x100>; > + interrupt-parent = <&intc>; You shouldn't need interrupt-parent here. In fact, I suspect it can be removed from all sibling nodes as well. Josh
On 10/18/2015 07:53 PM, Josh Cartwright wrote: > On Fri, Oct 16, 2015 at 03:42:29PM -0700, Moritz Fischer wrote: >> Added addtional nodes required for FPGA Manager operation >> of the Xilinx Zynq Devc configuration interface. >> >> Reviewed-by: Sören Brinkmann <soren.brinkmann@xilinx.com> >> Signed-off-by: Moritz Fischer <moritz.fischer@ettus.com> >> --- >> >> v2: No changes >> >> --- >> arch/arm/boot/dts/zynq-7000.dtsi | 5 +++++ >> 1 file changed, 5 insertions(+) >> >> diff --git a/arch/arm/boot/dts/zynq-7000.dtsi b/arch/arm/boot/dts/zynq-7000.dtsi >> index dc0457e..1a5220e 100644 >> --- a/arch/arm/boot/dts/zynq-7000.dtsi >> +++ b/arch/arm/boot/dts/zynq-7000.dtsi >> @@ -294,6 +294,11 @@ >> devcfg: devcfg@f8007000 { >> compatible = "xlnx,zynq-devcfg-1.0"; >> reg = <0xf8007000 0x100>; >> + interrupt-parent = <&intc>; > > You shouldn't need interrupt-parent here. In fact, I suspect it can be > removed from all sibling nodes as well. Correct. But I tend to do it vice-versa. To remove it from amba node and keep it in every IP here. The reason is simple to let everybody know that setting up right interrupt controller is something what they have to care. If you have more interrupt controllers in the system it can be messy. But again. Please remove this patch from this series. This will go to mainline through arm-soc but 1/3 and 3/3 will go through Greg. That's why please do not include it here. Thanks, Michal
On Mon, Oct 19, 2015 at 04:09:09PM +0200, Michal Simek wrote: > On 10/18/2015 07:53 PM, Josh Cartwright wrote: > > On Fri, Oct 16, 2015 at 03:42:29PM -0700, Moritz Fischer wrote: [..] > >> @@ -294,6 +294,11 @@ > >> devcfg: devcfg@f8007000 { > >> compatible = "xlnx,zynq-devcfg-1.0"; > >> reg = <0xf8007000 0x100>; > >> + interrupt-parent = <&intc>; > > > > You shouldn't need interrupt-parent here. In fact, I suspect it can be > > removed from all sibling nodes as well. > > Correct. But I tend to do it vice-versa. To remove it from amba node and > keep it in every IP here. The reason is simple to let everybody know > that setting up right interrupt controller is something what they have > to care. If you have more interrupt controllers in the system it can be > messy. If you like that, then you'd like the interrupt-extended properly even better (I think). > But again. Please remove this patch from this series. This will go to > mainline through arm-soc but 1/3 and 3/3 will go through Greg. That's > why please do not include it here. Not all of the patches in a series need to go through the same maintainer...splitting them up is common. It's nice seeing related things out on the list together. (Especially when there is a dependency at some level). Josh
On 10/19/2015 08:24 PM, Josh Cartwright wrote: > On Mon, Oct 19, 2015 at 04:09:09PM +0200, Michal Simek wrote: >> On 10/18/2015 07:53 PM, Josh Cartwright wrote: >>> On Fri, Oct 16, 2015 at 03:42:29PM -0700, Moritz Fischer wrote: > [..] >>>> @@ -294,6 +294,11 @@ >>>> devcfg: devcfg@f8007000 { >>>> compatible = "xlnx,zynq-devcfg-1.0"; >>>> reg = <0xf8007000 0x100>; >>>> + interrupt-parent = <&intc>; >>> >>> You shouldn't need interrupt-parent here. In fact, I suspect it can be >>> removed from all sibling nodes as well. >> >> Correct. But I tend to do it vice-versa. To remove it from amba node and >> keep it in every IP here. The reason is simple to let everybody know >> that setting up right interrupt controller is something what they have >> to care. If you have more interrupt controllers in the system it can be >> messy. > > If you like that, then you'd like the interrupt-extended properly even > better (I think). yep. That works for me too. Thanks, Michal
diff --git a/arch/arm/boot/dts/zynq-7000.dtsi b/arch/arm/boot/dts/zynq-7000.dtsi index dc0457e..1a5220e 100644 --- a/arch/arm/boot/dts/zynq-7000.dtsi +++ b/arch/arm/boot/dts/zynq-7000.dtsi @@ -294,6 +294,11 @@ devcfg: devcfg@f8007000 { compatible = "xlnx,zynq-devcfg-1.0"; reg = <0xf8007000 0x100>; + interrupt-parent = <&intc>; + interrupts = <0 8 4>; + clocks = <&clkc 12>; + clock-names = "ref_clk"; + syscon = <&slcr>; }; global_timer: timer@f8f00200 {