lkml.org 
[lkml]   [2015]   [Apr]   [21]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: [PATCH v5 3/4] documentation: dts: Add the device tree binding for APM X-Gene v1 PCIe MSI device tree node
On Tue, Apr 21, 2015 at 8:42 AM, Mark Rutland <mark.rutland@arm.com> wrote:
>
> On Tue, Apr 21, 2015 at 05:04:23AM +0100, Duc Dang wrote:
> > The driver for this binding is under 'drivers/pci/host/pci-xgene-msi.c'
>
> Please provide a bit of description of what this device is, and please
> place the binding patch _before_ the driver and DTS patches.

I will add in next version of the patch.
>
>
> > Signed-off-by: Duc Dang <dhdang@apm.com>
> > Signed-off-by: Tanmay Inamdar <tinamdar@apm.com>
> > ---
> > .../devicetree/bindings/pci/xgene-pci-msi.txt | 63 ++++++++++++++++++++++
> > 1 file changed, 63 insertions(+)
> > create mode 100644 Documentation/devicetree/bindings/pci/xgene-pci-msi.txt
> >
> > diff --git a/Documentation/devicetree/bindings/pci/xgene-pci-msi.txt b/Documentation/devicetree/bindings/pci/xgene-pci-msi.txt
> > new file mode 100644
> > index 0000000..0ffdcb3
> > --- /dev/null
> > +++ b/Documentation/devicetree/bindings/pci/xgene-pci-msi.txt
> > @@ -0,0 +1,63 @@
> > +* AppliedMicro X-Gene PCIe MSI interface
> > +
> > +Required properties:
> > +
> > +- compatible: should contain "apm,xgene1-msi" to identify the core.
>
> What does the core have to do with the MSI controller?

'core' here is the MSI controller block. I will use different word to
avoid confusion with CPU core.
>
> > +- msi-controller: indicates that this is X-Gene1 PCIe MSI controller node
> > +- reg: A list of physical base address and length for each set of controller
> > + registers.
>
> How many? Which ones? In which order? Do you need reg-names?

I will add these details in next version of the patch.
>
> > +- interrupts: A list of interrupt outputs of the controller.

I will add these details in next version of the patch.
>
> How many? Which ones? In which order? Do you need interrupt-names?

I will add these details in next version of the patch.
>
> You need to define these for *this particular binding*, in order for
> them to actually define the contract. An abstract definition is
> completely useless for writing or parsing a DT, and as such this
> document is just noise.
>
> Please think about the purpose of this document, and write something
> appropriate.
>
> [...]
>
> > + interrupts = < 0x0 0x10 0x4
> > + 0x0 0x11 0x4
> > + 0x0 0x12 0x4
> > + 0x0 0x13 0x4
> > + 0x0 0x14 0x4
> > + 0x0 0x15 0x4
> > + 0x0 0x16 0x4
> > + 0x0 0x17 0x4
> > + 0x0 0x18 0x4
> > + 0x0 0x19 0x4
> > + 0x0 0x1a 0x4
> > + 0x0 0x1b 0x4
> > + 0x0 0x1c 0x4
> > + 0x0 0x1d 0x4
> > + 0x0 0x1e 0x4
> > + 0x0 0x1f 0x4>;
>
> Nit: please bracket list entries individually.

Thanks for your comment, I will address them on next version of the patch.

>
> Mark.

Regards,
Duc Dang.


\
 
 \ /
  Last update: 2015-04-21 19:41    [W:0.094 / U:0.796 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site