Hi Rob,

On Wed, May 25, 2016 at 10:09 AM, Rob Herring [off-list ref] wrote:

On Mon, May 23, 2016 at 06:01:14PM -0700, Hoan Tran wrote:

quoted

Hi Rob,

Thanks for your review !

On Mon, May 23, 2016 at 1:30 PM, Rob Herring [off-list ref] wrote:

quoted

On Mon, May 16, 2016 at 09:17:25AM -0700, Hoan Tran wrote:

quoted

This patch adds the APM X-Gene hwmon device tree node documentation.

Signed-off-by: Hoan Tran <redacted>
---
 .../devicetree/bindings/hwmon/apm-xgene-hwmon.txt          | 14 ++++++++++++++
 1 file changed, 14 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/hwmon/apm-xgene-hwmon.txt

diff --git a/Documentation/devicetree/bindings/hwmon/apm-xgene-hwmon.txt b/Documentation/devicetree/bindings/hwmon/apm-xgene-hwmon.txt
new file mode 100644
index 0000000..49a482e
--- /dev/null
+++ b/Documentation/devicetree/bindings/hwmon/apm-xgene-hwmon.txt

@@ -0,0 +1,14 @@
+APM X-Gene hwmon driver
+
+Hwmon driver accesses sensors over the "SLIMpro" mailbox.

DT bindings describe h/w, not driver data.

How about this description: "APM X-Gene SOC sensors are accessed over
the "SLIMpro" mailbox" ?

quoted

I'm not sure this belongs in
DT and perhaps the devices for the mailbox should be created by the
mailbox driver.

I don't think the current mailbox supports it.

Then add it. The mailbox binding is really only needed when other h/w
blocks use a mailbox. As this is purely a firmware interface then the
mailbox driver can create any devices it needs. The devices don't have
to be in DT to be created.

Yes, I'll create a function inside mailbox. It allows client request a
channel by mailbox name and index without DT node.

Thanks
Hoan

Rob