Re: [PATCH 09/10] Input: synaptics-rmi4: Add device tree support to the SPI transport driver
From: Rob Herring <robh@kernel.org>
Date: 2015-11-30 21:28:08
Also in:
linux-devicetree, lkml
On Wed, Nov 25, 2015 at 04:10:25PM -0800, Andrew Duggan wrote:
quoted hunk ↗ jump to hunk
Add devicetree binding for SPI devices. Signed-off-by: Andrew Duggan <redacted> --- .../devicetree/bindings/input/rmi4/rmi_spi.txt | 57 ++++++++++++++++++++++ drivers/input/rmi4/rmi_spi.c | 44 ++++++++++++++++- 2 files changed, 100 insertions(+), 1 deletion(-) create mode 100644 Documentation/devicetree/bindings/input/rmi4/rmi_spi.txtdiff --git a/Documentation/devicetree/bindings/input/rmi4/rmi_spi.txt b/Documentation/devicetree/bindings/input/rmi4/rmi_spi.txt new file mode 100644 index 0000000..f20366b6 --- /dev/null +++ b/Documentation/devicetree/bindings/input/rmi4/rmi_spi.txt@@ -0,0 +1,57 @@ +Synaptics RMI4 SPI Device Binding + +The Synaptics RMI4 core is able to support RMI4 devices using differnet +transports and differnet functions. This file describes the device tree +bindings for devices using the SPI tranport driver. Complete documentation +for other transports and functions cen be found ini +Documentation/devicetree/bindings/input/rmi4. + +Required Properties: +- compatible: syna,rmi-spi +- reg: Chip select address for the device +- #address-cells: Set to 1 to indicate that the function child nodes + consist of only on uint32 value. +- #size-cells: Set to 0 to indicate that the function child nodes do not + have a size property. + +Optional Properties: +- interrupts: interrupt which the rmi device is connected to. +- interrupt-parent: The interrupt controller. +See Documentation/devicetree/bindings/interrupt-controller/interrupts.txt + +- syna,spi-read-delay: millisecond delay between read byte transfers. +- syna,spi-write-delay: millisecond delay between write byte transfers.
These could possibly be common SPI properties, but they would need to be somewhat better defined as every access on SPI is a read and a write. What if you have a read then write or write then read? Just a time between transfers would probably make more sense. Minimally, they should have -msec suffix. Rob