On 03/07/2016 12:33 PM, Jon Hunter wrote:
On 07/03/16 17:47, Stephen Warren wrote:
On 03/07/2016 03:00 AM, Jon Hunter wrote:
On 05/03/16 04:31, Rob Herring wrote:
On Fri, Mar 04, 2016 at 12:23:04PM +0000, Jon Hunter wrote:
Add power-domain binding documentation for the NVIDIA PMC driver in
order to support generic power-domains.
Signed-off-by: Jon Hunter <jonathanh@xxxxxxxxxx>
---
.../bindings/arm/tegra/nvidia,tegra20-pmc.txt | 80
++++++++++++++++++++++
1 file changed, 80 insertions(+)
diff --git
a/Documentation/devicetree/bindings/arm/tegra/nvidia,tegra20-pmc.txt
b/Documentation/devicetree/bindings/arm/tegra/nvidia,tegra20-pmc.txt
index 53aa5496c5cf..7d52bbe99709 100644
---
a/Documentation/devicetree/bindings/arm/tegra/nvidia,tegra20-pmc.txt
+++
b/Documentation/devicetree/bindings/arm/tegra/nvidia,tegra20-pmc.txt
@@ -1,5 +1,7 @@
NVIDIA Tegra Power Management Controller (PMC)
+== Power Management Controller Node ==
+
The PMC block interacts with an external Power Management Unit.
The PMC
mostly controls the entry and exit of the system from different sleep
modes. It provides power-gating controllers for SoC and CPU
power-islands.
@@ -70,6 +72,11 @@ Optional properties for hardware-triggered
thermal reset (inside 'i2c-thermtrip'
Defaults to 0. Valid values are described in
section 12.5.2
"Pinmux Support" of the Tegra4 Technical
Reference Manual.
+Optional nodes:
+- powergates : This node contains a hierarchy of power domain
nodes, which
+ should match the powergates on the Tegra SoC. See
"Powergate
+ Nodes" below.
+
Example:
/ SoC dts including file
@@ -115,3 +122,76 @@ pmc@7000f400 {
};
...
};
+
+
+== Powergate Nodes ==
+
+Each of the powergate nodes represent a power-domain on the Tegra SoC
+that can be power-gated by the Tegra PMC. The name of the powergate
node
+should be one of the below. Note that not every powergate is
applicable
+to all Tegra devices and the following list shows which powergates are
+applicable to which devices. Please refer to the Tegra TRM for more
+details on the various powergates.
+
+ Name Description Devices Applicable
+ 3d 3D Graphics Tegra20/114/124/210
+ 3d0 3D Graphics 0 Tegra30
+ 3d1 3D Graphics 1 Tegra30
+ aud Audio Tegra210
+ dfd Debug Tegra210
+ dis Display A Tegra114/124/210
+ disb Display B Tegra114/124/210
+ heg 2D Graphics Tegra30/114/124/210
+ iram Internal RAM Tegra124/210
+ mpe MPEG Encode All
+ nvdec NVIDIA Video Decode Engine Tegra210
+ nvjpg NVIDIA JPEG Engine Tegra210
+ pcie PCIE Tegra20/30/124/210
+ sata SATA Tegra30/124/210
+ sor Display interfaces Tegra124/210
+ ve2 Video Encode Engine 2 Tegra210
+ venc Video Encode Engine All
+ vdec Video Decode Engine Tegra20/30/114/124
+ vic Video Imaging Compositor Tegra124/210
+ xusba USB Partition A Tegra114/124/210
+ xusbb USB Partition B Tegra114/124/210
+ xusbc USB Partition C Tegra114/124/210
+
+Required properties:
+ - clocks: Must contain an entry for each clock required by the
PMC for
+ controlling a power-gate. See ../clocks/clock-bindings.txt for
details.
+ - resets: Must contain an entry for each reset required by the
PMC for
+ controlling a power-gate. See ../reset/reset.txt for details.
+ - #power-domain-cells: Must be 0.
+
+Example:
+
+ pmc: pmc@0,7000e400 {
Just pmc@7000e400. We were erronously using commas for 64-bit addresses
briefly, but the correct usage is when there are distinct fields in reg
like PCI bus,dev,func.
Ok. Looks like we use this style quite widely across all Tegra 32-bit
and 64-bit devices for all peripheral devices.
Thierry, Stephen,
Do we need to correct this for existing devices?
I don't think so; my understanding is that where reg has multiple
address cells, they all get represented in the unit address. Otherwise,
the node names couldn't be guaranteed unique.
OK, but I am a bit confused as that does not sound exactly the same as
what Rob is saying?
Consider a SoC with two I2C ports at address 1_00000000 and 2_00000000
physical. The top 32-bits (2nd cell) of the address must be represented
in the unit address, or the nodes won't have a unique name.
Perhaps Rob is simply saying that multi-cell values should be treated as
a single large integer, rather than comma-separating each cell value? I
perhaps mistakenly took his comment to mean that only a single address
cell should be used in the unit address.
Also, I see that starting with tegra124 we have #address-cells = <2> and
#size-cells = <2>, which I also don't understand either. Are the bus
addresses really 64-bit for tegra124?
T124 has a >32-bit bus since it supports >2GB RAM (via LPAE) (although
not a full 64-bit bus I would imagine). That alone means we need two
cells in DT to represent large RAM layouts.
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
