On Thu, Mar 4, 2021 at 12:03 AM Serge Semin <fancer.lancer@xxxxxxxxx> wrote:
>
> On Wed, Mar 03, 2021 at 07:41:40PM -0800, Brad Larson wrote:
> > Add Pensando common and Elba SoC specific device nodes
> > and corresponding binding documentation.
>
> This also needs to be split up into sub-patches seeing these are
> unrelated changes like device bindings update, new platform DT file.
In patchset v2 this is split into sub-patches.
> What about converting this file to DT-schema and adding new HW
> bindings in there?
Converted existing file devicetree/bindings/spi/cadence-quadspi.txt to
YAML schema.
> > +&spi0 {
> > + num-cs = <4>;
>
> > + cs-gpios = <&spics 0 0>, <&spics 1 0>, <&porta 1 0>, <&porta 7 0>;
>
> Oh, you've got four peripheral SPI devices connected with only two native CS
> available. Hmm, then I don't really know a better way, but just to forget about
> the native DW APB CS functionality and activate the direct driving of
> all the CS-pins at the moment of the DW APB SPI controller probe
> procedure. Then indeed you'll need a custom CS function defined in the DW APB
> SPI driver to handle that.
Yes, with an Elba SoC specific gpio driver.
> So that GPIO-controller is just a single register which provides a way
> to toggle the DW APB SPI CS-mode together with their output value.
> If so and seeing there are a few more tiny spaces of config
> registers added to eMMC, PCI, etc DT node, I suppose all of them
> belong to some bigger config space of the SoC. Thus I'd suggest to at
> least implement them as part of a System Controller DT node. Then use
> that device service to switch on/off corresponding functionality.
> See [2] and the rest of added to the kernel DTS files with
> syscon-nodes for example.
>
> [2] Documentation/devicetree/bindings/mfd/syscon.yaml
To us it was more understandable to implement a standard gpio driver
for the spi chip-selects.
