Add support for GE SUNH hot-pluggable connector
Subsystems (8)
Luca Ceresoli
reviewerv3 cover Add support for GE SUNH hot-pluggable connector
Hello,
this series aims at supporting a Linux device with a connector to physically add and remove an add-on to/from the main device to augment its features at runtime, using device tree overlays.
What's new in v3 (a lot!)
=========================
There are big changes compared to v2, in many areas. Here are the main ones.
* Totally new device tree structure, much more rational as recommended by Rob. Now overlays are fully decoupled from the connector and adding properties to nodes in the base tree is (almost) not needed anymore. Read below for a description of the new format.
* The DRM hotplug-bridge driver now adds and removes dynamically a DRM connector on hot(un)plug, correctly representing the LVDS connector that comes and goes.
However a few new rough edges emerged that are not yet solved in this v3. Discussion would help in finding the right direction:
* Describing the NVMEM cell addition still requires adding two properties to a node in the base tree. Not sure the current NVMEM cell bindings allow to do better.
* I2C is one of the non-discoverable busses and I added a patch to i2c-core-of.c to allow it to handle the new format. It works, but with two caveats, see patch 5 for the details.
* Sima proposed various improvements to the DRM hotplug-driver. For now I only did the one that has userspace impact (dynamic creation and removal of a connector). The others are still pending, and not all are clear to me, so I'd be glad to discuss this again as part of this series.
* An issue has emerged with the devlink usage in the regulator subsystem. I analyzed the problem but the solution I found may well be wrong. Details in patch 7.
Now back to introducing the series.
Use case
========
This series targets a professional product (GE SUNH) that is composed of a "main" part running on battery, with the main SoC and able to work autonomously with limited features, and an optional "add-on" that enables more features by adding more hardware peripherals, some of which are on non-discoverable busses such as I2C and MIPI DSI.
The add-on can be connected and disconnected at runtime at any moment by the end user, and add-on features need to be enabled and disabled automatically at runtime.
The add-on has status pins that are connected to GPIOs on the main board, allowing the CPU to detect add-on insertion and removal. It also has a reset GPIO allowing to reset all peripherals on the add-on at once.
The features provided by the add-on include a display and a battery charger to recharge the battery of the main part. The display on the add-on has an LVDS input but the connector between the base and the add-on has a MIPI DSI bus, so a DSI-to-LVDS bridge is present on the add-on.
Different add-on models can be connected to the main part, and for this a model ID is stored in the add-on itself so the software running on the CPU on the main part knows which non-discoverable hardware to probe.
Overall approach
================
Device tree overlays appear as the most natural solution to support the addition and removal of devices from a running system.
Several features are missing from the mainline Linux kernel in order to support this use case:
1. runtime (un)loading of device tree overlays is currently not exposed
2. if enabled, overlay (un)loading exposes several issues and limitations
3. the DRM subsystem assumes video bridges are non-removable
This series targets items 1 and 3 and some of the issues mentioned in item
2. Other issues are being handled separately (see "Device tree overlay issues" below).
Device tree representation and connector driver
===============================================
The device tree description we propose involves 3 main parts.
1: the main (fixed) device tree
The main device tree describes the connector itself along with the status and reset GPIOs. It also provides a 'ports' node to describe how the two sides of the MIPI DSI bus connect to each other. So now, differently from v1, there is no standalone representation of the DRM bridge because it is not really a hardware component. Here is how the connector is represented in the fixed part of the device tree:
/ {
#include <dt-bindings/gpio/gpio.h>
addon_connector: addon-connector {
compatible = "ge,sunh-addon-connector";
reset-gpios = <&gpio1 1 GPIO_ACTIVE_LOW>;
plugged-gpios = <&gpio1 2 GPIO_ACTIVE_LOW>;
powergood-gpios = <&gpio1 3 GPIO_ACTIVE_HIGH>;
i2c-dbat {
i2c-parent = <&i2c2_ch1>;
#address-cells = <1>;
#size-cells = <0>;
};
i2c-gp {
i2c-parent = <&i2c5>;
#address-cells = <1>;
#size-cells = <0>;
};
i2c-btp {
i2c-parent = <&i2c3>;
#address-cells = <1>;
#size-cells = <0>;
};
dsi {
ports {
#address-cells = <1>;
#size-cells = <0>;
port@0 {
reg = <0>;
hotplug_bridge_sink: endpoint {
remote-endpoint = <&dsi_to_hotplug_bridge>;
};
};
};
};
};
};The connector has a specific compatible string, and this series adds a driver supporting it. This driver uses the device tree overlay loading and unloading facilities already implemented by the kernel but not currently exposed.
The driver detects the connection status from the GPIOs and reacts to a connection event by loading a first overlay (the "base" overlay).
The 'i2c-*' nodes represent the hot-pluggable section of I2C busses crossing the connector, whose controller is on the main board but which have devices on the add-on. There is one node per each such bus: this allows full decoupling between the base board and the overlay, but requires an additional info to associate the 'i2c-*' node to the physical bus controller.
Similarly, the 'dsi' node contains the fixed part of the video pipeline. 'port@1' represents the physical bus that terminates on the connector. 'port@1', representing the continuation of the video bus in the add-on, is added by the overlay.
The 'dsi' node would also allow to describe a similar connector having multiple video busses: these would have one node each, such as 'dsi-foo', 'dsi-bar', 'lvds-foo', 'lvds-bar' etc while keeping the ports for each connector appropriately separated.
2: the "base" overlay
The "base" overlay describes the common components that are required to read the model ID. These are identical for all add-on models, thus only one "base" overlay is needed:
/dts-v1/;
/plugin/;
/ {
fragment@0 {
target-path = "";
__overlay__ {
nvmem-cells = <&addon_id>;
nvmem-cell-names = "id";
i2c-dbat {
addon_eeprom: eeprom@51 {
compatible = "atmel,24c64";
reg = <0x51>;
pagesize = <32>;
nvmem-layout {
compatible = "fixed-layout";
#address-cells = <1>;
#size-cells = <1>;
/* Data cells */
addon_id: addon-id@400 {
reg = <0x400 0x1>;
};
};
};
};
};
};
};Note the overlay does not have a target node, unlike the v2 series. This allows the overlay to be fully decoupled from the base tree, and conceptually to have multiple compatible connectors on the same board. It also avoids the need to add properties to nodes in the base tree, with an exception.
Indeed this v3 the 'nvmem-cells' and 'nvmem-cell-names' are the only two properties added to a node that is in the base tree. This is still waiting for a different representation to avoid adding such properties and all the deadprops and leaks thereof.
Here an I2C device is added by a subnode of 'i2c-dbat' for an EEPROM. The i2c-dbat node itself is already present in the base tree, carrying the link to the actual I2C adapter node.
The EEPROM holds the model ID of each add-on, using always the same I2C address and memory offset.
3: the "add-on-specific" overlay
Based on the model ID, the connector driver loads the second overlay, which describes all the add-on hardware not yet described by the base overlay. This overlay is model-specific.
Excerpt:
/ {
fragment@0 {
target-path = "";
__overlay__ {
dsi {
ports {
port@1 {
reg = <1>;
hotplug_bridge_source: endpoint {
remote-endpoint = <&sn65dsi84_from_bridge>;
};
};
};
};
i2c-gp {
#address-cells = <1>;
#size-cells = <0>;
dsi-lvds-bridge@2c {
compatible = "ti,sn65dsi84";
reg = <0x2c>;
ports {
port@0 {
reg = <0>;
sn65dsi84_from_bridge: endpoint {
remote-endpoint = <&hotplug_bridge_source>;
data-lanes = <1 2 3 4>;
};
};
port@2 {
reg = <2>;
sn65dsi84_out0: endpoint {
remote-endpoint = <&panel_dsi_lvds_in0>;
};
};
port@3 {
reg = <3>;
sn65dsi84_out1: endpoint {
remote-endpoint = <&panel_dsi_lvds_in1>;
};
};
};
};
};
nobus-devices {
reg_addon_3v3_lcd: regulator-addon-3v3-lcd {
compatible = "regulator-fixed";
regulator-name = "3V3_LCD_ADDON";
...
};
backlight_addon: backlight-addon {
compatible = "led-backlight";
...
};
addon_panel_dsi_lvds: panel-dsi-lvds {
compatible = "...";
power-supply = <®_addon_3v3_lcd>;
backlight = <&backlight_addon>;
ports {
#address-cells = <1>;
#size-cells = <0>;
port@0{
reg = <0>;
dual-lvds-odd-pixels;
panel_dsi_lvds_in0: endpoint {
remote-endpoint = <&sn65dsi84_out0>;
};
};
port@1{
reg = <1>;
dual-lvds-even-pixels;
panel_dsi_lvds_in1: endpoint {
remote-endpoint = <&sn65dsi84_out1>;
};
};
};
};
};
};
};
};Here the 'dsi/ports/port@1' node is completing the 'dsi' section already present in the base tree, thus describing that this add-on is connecting those DSI lines to something, in this case the DSI-to-LVDS bridge and a panel.
The 'nobus-devices' node, which is also new in this v3, containts one subnode for each device that is not on any CPU-reachable bus (I2C, DSI, etc): fixed/GPIO regulators, backlight, the panel etc. Nodes for these devices are children of the root node in normal (no overlay) device tree systems, and are probed as platform devices by kernel code. With the connector we need to have them under the connector node, and the choice was to let them be children of the connector node or to group them into a new subnode. We chose the latter because that provides a more explicit representation of reality, and is coherent with the 'dsi' and 'i2c-*' nodes. As a good side effect for the implementation, this means other nodes under the connector node (dsi, i2c-*) are not considered when populating platform devices.
After these steps, the add-on is fully described and working on the system. When the status GPIOs report a disconnection, the overlays are unloaded in reverse order.
DRM hotplug bridge driver
=========================
DRM natively supports pipelines whose display can be removed, but all the components preceding it (all the display controller and any bridges) are assumed to be fixed and cannot be plugged, removed or modified at runtime.
This series adds support for DRM pipelines having a removable part after the encoder, thus also allowing bridges to be removed and reconnected at runtime, possibly with different components.
This picture summarizes the DRM structure implemented by this series:
.------------------------.
| DISPLAY CONTROLLER |
| .---------. .------. |
| | ENCODER |<--| CRTC | |
| '---------' '------' |
'------|-----------------'
|
|DSI HOTPLUG
V CONNECTOR
.---------. .--. .-. .---------. .-------.
| 0 to N | | _| _| | | 1 to N | | |
| BRIDGES |--DSI-->||_ |_ |--DSI-->| BRIDGES |--LVDS-->| PANEL |
| | | | | | | | | |
'---------' '--' '-' '---------' '-------'[--- fixed components --] [----------- removable add-on -----------]
Fixed components include:
* all components up to the DRM encoder, usually part of the SoC
* optionally some bridges, in the SoC and/or as external chips
Components on the removable add-on include:
* one or more bridges
* a fixed connector (not one natively supporting hotplug such as HDMI)
* the panel
The video bus is MIPI DSI in the example and in the implementation provided by this series, but the implementation is meant to allow generalization to other video busses without native hotplug support, such as parallel video and LVDS.
Note that the term "connector" in this context is different from the "DRM connector" abstraction already present in the DRM subsystem (struct drm_connector).
More details in the commit message of patch 4.
That's all
==========
Thanks for you patience in reading this!
Luca
Too many changes in v3 to mention them all, but here are the big ones:
- Rewrote the DT format to allow fully decoupled overlays and to avoid
adding properties (with the NVMEM exception still to be solved)
- Implemented device instantiation based on the new DT format: i2c in i2c-core-of.c nobus-devices in the connector driver
- DRM: insert/remove an LVDS DRM connector on hot(un)plug events
- Added patch for a devlink issue on overlay removal (mostly to start
discussion)
- Link to v2: https://lore.kernel.org/r/[email protected]
Changes in v2:
- Added bindings and driver for ge,sunh-addon-connector
- Removed bindings for the hotplug-video-connector, this is now represented
in DT as part of the ge,sunh-addon-connector
- Various monior improvements to the DRM hotplug-bridge driver
- Link to v1: https://lore.kernel.org/r/[email protected]
---
Luca Ceresoli (6):
dt-bindings: connector: add GE SUNH hotplug addon connector drm/encoder: add drm_encoder_cleanup_from()
drm/bridge: hotplug-bridge: add driver to support hot-pluggable DSI bridges
i2c: i2c-core-of: follow i2c-parent phandle to probe devices from added nodes
misc: add ge-addon-connector driver
[DO NOT APPLY] driver core: do not unblock consumers any drivers found
Paul Kocialkowski (1):
drm/bridge: add bridge notifier to be notified of bridge addition and removal
.../connector/ge,sunh-addon-connector.yaml | 185 ++++++ MAINTAINERS | 11 + drivers/base/core.c | 21 - drivers/gpu/drm/bridge/Kconfig | 15 + drivers/gpu/drm/bridge/Makefile | 1 + drivers/gpu/drm/bridge/hotplug-bridge.c | 665 +++++++++++++++++++++ drivers/gpu/drm/drm_bridge.c | 35 ++ drivers/gpu/drm/drm_encoder.c | 21 + drivers/i2c/i2c-core-of.c | 9 + drivers/misc/Kconfig | 18 + drivers/misc/Makefile | 1 + drivers/misc/ge-sunh-connector.c | 523 ++++++++++++++++ include/drm/drm_bridge.h | 19 + include/drm/drm_encoder.h | 1 + 14 files changed, 1504 insertions(+), 21 deletions(-)
---
base-commit: 5c2a33cdaccfa6ecc787f00fb8f034ffb03706be
change-id: 20240319-hotplug-drm-bridge-16b86e67fe92
Best regards,
Signature · 1 line
Luca Ceresoli <[email protected]>
Luca Ceresoli
reviewerPATCH v3 1/7 dt-bindings: connector: add GE SUNH hotplug addon connector
Add bindings for the GE SUNH add-on connector. This is a physical, hot-pluggable connector that allows to attach and detach at runtime an add-on adding peripherals on non-discoverable busses.
---
Changed in v3:
- change the layout to only add subnodes, not properties
- add the 'nobus-devices' node description to hold devices not on any bus
- add 'i2c-*' nodes for the I2C busses, using a i2c-parent phandle
- and the 'dsi' node for the DSI bus
- move the entire port@1 node to the overlay (not only the remote-endpoint property)
- remove the overlay examples (Overlays in examples are not supported)
- add more clarifying descriptions and comments for examples
- some rewording
This patch was added in v2.
--- .../connector/ge,sunh-addon-connector.yaml | 185 +++++++++++++++++++++ MAINTAINERS | 5 + 2 files changed, 190 insertions(+)
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml new file +185 -0
@@ -0,0 +1,185 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/connector/ge,sunh-addon-connector.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: GE SUNH hotplug add-on connector + +maintainers: + - Luca Ceresoli <[email protected]> + +description: + Represent the physical connector present on GE SUNH devices that allows + to attach and detach at runtime an add-on adding peripherals on + non-discoverable busses. Peripherals on the add-on include I2C sensors + and a video bridge controlled via I2C. + + The connector has status GPIOs to notify the connection status to the CPU + and a reset GPIO to allow the CPU to reset all the peripherals on the + add-on. It also has I2C busses and a 4-lane MIPI DSI bus. + + Different add-on models can be connected, each having different + peripherals. For this reason each add-on has a model ID stored in a + non-volatile memory, which is accessed in the same way on all add-ons. + + Add-on removal can happen at any moment under user control and without + prior notice to the CPU, making all of its components not usable + anymore. Later on, the same or a different add-on model can be connected. + +properties: + compatible: + const: ge,sunh-addon-connector + + reset-gpios: + description: An output GPIO to reset the peripherals on the add-on. + maxItems: 1 + + plugged-gpios: + description: An input GPIO that is asserted if and only if an add-on is + physically connected. + maxItems: 1 + + powergood-gpios: + description: An input GPIO that is asserted if and only if power rails + on the add-on are stable. + maxItems: 1 + + nobus-devices: + description: + A container for devices not accessible via any data bus. Common use + cases include fixed and GPIO regulators, simple video panels and LED + or GPIO backlight devices. When not hot-pluggable, nodes such devices + are children of the root node. + + This node should not be present in the connector description in the + base device tree. It should be added by overlays along with a subnode + per device. + + type: object + additionalProperties: false + + dsi: + type: object + additionalProperties: false + + properties: + ports: + $ref: /schemas/graph.yaml#/properties/ports + + description: + OF graph bindings modeling the MIPI DSI bus across the connector. The + connector splits the video pipeline in a fixed part and a removable + part. + + The fixed part of the video pipeline includes all components up to + the display controller and 0 or more bridges. The removable part + includes any bridges and any other components up to the panel. + + properties: + port@0: + $ref: /schemas/graph.yaml#/properties/port + description: + The last point of the non-removable part of the MIPI DSI bus + line. The remote-endpoint sub-node must point to the last + non-removable video component of the video pipeline. + + port@1: + $ref: /schemas/graph.yaml#/properties/port + description: + The first point of the removable part of the MIPI DSI bus + line. The remote-endpoint sub-node must point to the first + video pipeline component on the add-on. As it describes the + hot-pluggable hardware, the endpoint node cannot be filled + until an add-on is detected, so this node needs to be added + by a device tree overlay at runtime. + + required: + - port@0 + # port@1 is added by the overlay for any add-on using the DSI lines + + required: + - ports + +patternProperties: + '^i2c-(dbat|gp|btp)$': + description: + An I2C bus that goes through the connector. The adapter (and possibly + some clients) are on the fixed side. Add-ons that have any clients on + this bus have to be added by the add-on overlay, inside this node. + + $ref: /schemas/i2c/i2c-controller.yaml# + unevaluatedProperties: false + type: object + + properties: + i2c-parent: + $ref: /schemas/types.yaml#/definitions/phandle + description: + Phandle pointing to the I2C bus controller on the fixed side that + drives this bus + +required: + - compatible + - i2c-dbat + - i2c-gp + - i2c-btp + - dsi + +unevaluatedProperties: false + +examples: + # This is the description of the connector as it should appear in the + # main DTS describing the "main" board up to the connector. This is + # supposed to be used together with the overlays in the two following + # examples. The addon_connector and hotplug_conn_dsi_out labels are + # referenced by the overlays in those examples. + - | + / { + #include <dt-bindings/gpio/gpio.h> + + addon_connector: addon-connector { + compatible = "ge,sunh-addon-connector"; + reset-gpios = <&gpio1 1 GPIO_ACTIVE_LOW>; + plugged-gpios = <&gpio1 2 GPIO_ACTIVE_LOW>; + powergood-gpios = <&gpio1 3 GPIO_ACTIVE_HIGH>; + + i2c-dbat { + i2c-parent = <&i2c5_ch2>; + #address-cells = <1>; + #size-cells = <0>; + // device subnodes to be added by overlays + }; + + i2c-gp { + i2c-parent = <&i2c4>; + #address-cells = <1>; + #size-cells = <0>; + // device subnodes to be added by overlays + }; + + i2c-btp { + i2c-parent = <&i2c3>; + #address-cells = <1>; + #size-cells = <0>; + // device subnodes to be added by overlays + }; + + dsi { + ports { + #address-cells = <1>; + #size-cells = <0>; + + port@0 { + reg = <0>; + + endpoint { + remote-endpoint = <&previous_bridge_out>; + }; + }; + + // port@1 to be added by overlay + }; + }; + }; + };
MAINTAINERS +5 -0
@@ -10254,6 +10254,11 @@ S: Maintained F: Documentation/devicetree/bindings/iio/pressure/honeywell,mprls0025pa.yaml F: drivers/iio/pressure/mprls0025pa* +HOTPLUG CONNECTOR FOR GE SUNH ADDONS +M: Luca Ceresoli <[email protected]> +S: Maintained +F: Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml + HP BIOSCFG DRIVER M: Jorge Lopez <[email protected]> L: [email protected]
Signature · 1 line
2.34.1
Rob Herring (Arm)
maintainerRe: [PATCH v3 1/7] dt-bindings: connector: add GE SUNH hotplug addon connector
Quoted 19 lines · Luca Ceresoli
Add bindings for the GE SUNH add-on connector. This is a physical, hot-pluggable connector that allows to attach and detach at runtime an add-on adding peripherals on non-discoverable busses.
Signed-off-by Luca Ceresoli <[email protected]>---
Changed in v3:
- change the layout to only add subnodes, not properties
- add the 'nobus-devices' node description to hold devices not on any bus
- add 'i2c-*' nodes for the I2C busses, using a i2c-parent phandle
- and the 'dsi' node for the DSI bus
- move the entire port@1 node to the overlay (not only the remote-endpoint property)
- remove the overlay examples (Overlays in examples are not supported)
- add more clarifying descriptions and comments for examples
- some rewording
Quoted 6 lines
This patch was added in v2.
--- .../connector/ge,sunh-addon-connector.yaml | 185 +++++++++++++++++++++ MAINTAINERS | 5 + 2 files changed, 190 insertions(+)
My bot found errors running 'make dt_binding_check' on your patch:
yamllint warnings/errors:
dtschema/dtc warnings/errors:
/builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.example.dtb: /: 'compatible' is a required property
from schema $id: http://devicetree.org/schemas/root-node.yaml#
/builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.example.dtb: /: 'model' is a required property
from schema $id: http://devicetree.org/schemas/root-node.yaml#
/builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.example.dtb: /: '#address-cells' is a required property
from schema $id: http://devicetree.org/schemas/root-node.yaml#
/builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.example.dtb: /: '#size-cells' is a required property
from schema $id: http://devicetree.org/schemas/root-node.yaml#
doc reference errors (make refcheckdocs):
See https://patchwork.ozlabs.org/project/devicetree-bindings/patch/[email protected]
The base for the series is generally the latest rc1. A different dependency should be noted in *this* patch.
If you already ran 'make dt_binding_check' and didn't see the above error(s), then make sure 'yamllint' is installed and dt-schema is up to date:
pip3 install dtschema --upgrade
Please check and re-submit after running the above command yourself. Note that DT_SCHEMA_FILES can be set to your schema file to speed up checking your schema. However, it must be unset to test all examples with your schema.
Luca Ceresoli
reviewerRe: [PATCH v3 1/7] dt-bindings: connector: add GE SUNH hotplug addon connector
In reply to Luca CeresoliHello,
Quoted 5 lines · On Fri, 09 Aug 2024 17:34:49 +0200 Luca Ceresoli <[email protected]>…
Add bindings for the GE SUNH add-on connector. This is a physical, hot-pluggable connector that allows to attach and detach at runtime an add-on adding peripherals on non-discoverable busses.
Signed-off-by Luca Ceresoli <[email protected]>
...
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml
Quoted 3 lines
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml new file +3
-0+examples: + # This is the description of the connector as it should appear in the + # main DTS describing the "main" board up to the connector.
The comment from now on has outdated content referring to v2 and which I forgot to remove before sending v3:
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml line 136
Quoted 4 lines
This is
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml new file +3
-0+ # supposed to be used together with the overlays in the two following + # examples. The addon_connector and hotplug_conn_dsi_out labels are + # referenced by the overlays in those examples.
Fixed locally.
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml line 139
Quoted 3 lines
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml new file +3
-0+ - | + / { + #include <dt-bindings/gpio/gpio.h>
I also removed the root node which is not needed and cause bot warnings, and with that done the #include went back to a correct position.
The above are queued for v4.
Luca
Signature · 3 lines
Luca Ceresoli, Bootlin Embedded Linux and Kernel engineering https://bootlin.com
Rob Herring
maintainerRe: [PATCH v3 1/7] dt-bindings: connector: add GE SUNH hotplug addon connector
In reply to Luca CeresoliQuoted 3 lines · Luca Ceresoli
Add bindings for the GE SUNH add-on connector. This is a physical, hot-pluggable connector that allows to attach and detach at runtime an add-on adding peripherals on non-discoverable busses.
Overall, looks pretty good.
Quoted 30 lines
Signed-off-by Luca Ceresoli <[email protected]>---
Changed in v3:
- change the layout to only add subnodes, not properties
- add the 'nobus-devices' node description to hold devices not on any bus
- add 'i2c-*' nodes for the I2C busses, using a i2c-parent phandle
- and the 'dsi' node for the DSI bus
- move the entire port@1 node to the overlay (not only the remote-endpoint property)
- remove the overlay examples (Overlays in examples are not supported)
- add more clarifying descriptions and comments for examples
- some rewordingThis patch was added in v2.
--- .../connector/ge,sunh-addon-connector.yaml | 185 +++++++++++++++++++++ MAINTAINERS | 5 + 2 files changed, 190 insertions(+)Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml new file +2
-0@@ -0,0 +1,185 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml line 12
Quoted 10 lines
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml new file +10
-0+--- +$id: http://devicetree.org/schemas/connector/ge,sunh-addon-connector.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: GE SUNH hotplug add-on connector + +maintainers: + - Luca Ceresoli <[email protected]> + +description:
You need '|' or '>' to preserve line breaks. Elsewhere too.
Quoted 13 lines
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml new file +13
-0+ Represent the physical connector present on GE SUNH devices that allows + to attach and detach at runtime an add-on adding peripherals on + non-discoverable busses. Peripherals on the add-on include I2C sensors + and a video bridge controlled via I2C. + + The connector has status GPIOs to notify the connection status to the CPU + and a reset GPIO to allow the CPU to reset all the peripherals on the + add-on. It also has I2C busses and a 4-lane MIPI DSI bus. + + Different add-on models can be connected, each having different + peripherals. For this reason each add-on has a model ID stored in a + non-volatile memory, which is accessed in the same way on all add-ons. +
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml line 35
Quoted 10 lines
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml new file +10
-0+ Add-on removal can happen at any moment under user control and without + prior notice to the CPU, making all of its components not usable + anymore. Later on, the same or a different add-on model can be connected. + +properties: + compatible: + const: ge,sunh-addon-connector + + reset-gpios: + description: An output GPIO to reset the peripherals on the add-on.
Active state is?
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml line 40
Quoted 5 lines
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml new file +5
-0+ maxItems: 1 + + plugged-gpios: + description: An input GPIO that is asserted if and only if an add-on is + physically connected.
Asserted is high or low?
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml line 45
Quoted 5 lines
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml new file +5
-0+ maxItems: 1 + + powergood-gpios: + description: An input GPIO that is asserted if and only if power rails + on the add-on are stable.
Active state is?
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml line 48
Quoted 3 lines
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml new file +3
-0+ maxItems: 1 + + nobus-devices:
Just 'devices'.
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml line 60
Quoted 12 lines
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml new file +12
-0+ description: + A container for devices not accessible via any data bus. Common use + cases include fixed and GPIO regulators, simple video panels and LED + or GPIO backlight devices. When not hot-pluggable, nodes such devices + are children of the root node. + + This node should not be present in the connector description in the + base device tree. It should be added by overlays along with a subnode + per device. + + type: object + additionalProperties: false
The schema needs to work with the overlay applied too. 'true' is fine here.
Quoted 145 lines
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml new file +125
-0+ + dsi: + type: object + additionalProperties: false + + properties: + ports: + $ref: /schemas/graph.yaml#/properties/ports + + description: + OF graph bindings modeling the MIPI DSI bus across the connector. The + connector splits the video pipeline in a fixed part and a removable + part. + + The fixed part of the video pipeline includes all components up to + the display controller and 0 or more bridges. The removable part + includes any bridges and any other components up to the panel. + + properties: + port@0: + $ref: /schemas/graph.yaml#/properties/port + description: + The last point of the non-removable part of the MIPI DSI bus + line. The remote-endpoint sub-node must point to the last + non-removable video component of the video pipeline. + + port@1: + $ref: /schemas/graph.yaml#/properties/port + description: + The first point of the removable part of the MIPI DSI bus + line. The remote-endpoint sub-node must point to the first + video pipeline component on the add-on. As it describes the + hot-pluggable hardware, the endpoint node cannot be filled + until an add-on is detected, so this node needs to be added + by a device tree overlay at runtime. + + required: + - port@0 + # port@1 is added by the overlay for any add-on using the DSI lines + + required: + - ports + +patternProperties: + '^i2c-(dbat|gp|btp)$': + description: + An I2C bus that goes through the connector. The adapter (and possibly + some clients) are on the fixed side. Add-ons that have any clients on + this bus have to be added by the add-on overlay, inside this node. + + $ref: /schemas/i2c/i2c-controller.yaml# + unevaluatedProperties: false + type: object + + properties: + i2c-parent: + $ref: /schemas/types.yaml#/definitions/phandle + description: + Phandle pointing to the I2C bus controller on the fixed side that + drives this bus + +required: + - compatible + - i2c-dbat + - i2c-gp + - i2c-btp + - dsi + +unevaluatedProperties: false + +examples: + # This is the description of the connector as it should appear in the + # main DTS describing the "main" board up to the connector. This is + # supposed to be used together with the overlays in the two following + # examples. The addon_connector and hotplug_conn_dsi_out labels are + # referenced by the overlays in those examples. + - | + / { + #include <dt-bindings/gpio/gpio.h> + + addon_connector: addon-connector { + compatible = "ge,sunh-addon-connector"; + reset-gpios = <&gpio1 1 GPIO_ACTIVE_LOW>; + plugged-gpios = <&gpio1 2 GPIO_ACTIVE_LOW>; + powergood-gpios = <&gpio1 3 GPIO_ACTIVE_HIGH>; + + i2c-dbat { + i2c-parent = <&i2c5_ch2>; + #address-cells = <1>; + #size-cells = <0>; + // device subnodes to be added by overlays + }; + + i2c-gp { + i2c-parent = <&i2c4>; + #address-cells = <1>; + #size-cells = <0>; + // device subnodes to be added by overlays + }; + + i2c-btp { + i2c-parent = <&i2c3>; + #address-cells = <1>; + #size-cells = <0>; + // device subnodes to be added by overlays + }; + + dsi { + ports { + #address-cells = <1>; + #size-cells = <0>; + + port@0 { + reg = <0>; + + endpoint { + remote-endpoint = <&previous_bridge_out>; + }; + }; + + // port@1 to be added by overlay + }; + }; + }; + };MAINTAINERS +5
-0@@ -10254,6 +10254,11 @@ S: Maintained F: Documentation/devicetree/bindings/iio/pressure/honeywell,mprls0025pa.yaml F: drivers/iio/pressure/mprls0025pa* +HOTPLUG CONNECTOR FOR GE SUNH ADDONS +M: Luca Ceresoli <[email protected]> +S: Maintained +F: Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml + HP BIOSCFG DRIVER M: Jorge Lopez <[email protected]> L: [email protected]Signature · 2 lines
-- 2.34.1
Luca Ceresoli
reviewerRe: [PATCH v3 1/7] dt-bindings: connector: add GE SUNH hotplug addon connector
Hello Rob,
Quoted 6 lines · On Tue, 13 Aug 2024 09:19:01 -0600 Rob Herring <[email protected]> wrote:
Quoted 3 lines · Luca Ceresoli
Add bindings for the GE SUNH add-on connector. This is a physical, hot-pluggable connector that allows to attach and detach at runtime an add-on adding peripherals on non-discoverable busses.
Overall, looks pretty good.
Thanks, I'm very glad it does.
Quoted 5 lines
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml line 48
Quoted 3 lines
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml new file +3
-0+ maxItems: 1 + + nobus-devices:Just 'devices'.
Sure, simple enough.
Quoted 7 lines
Quoted 7 lines
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml new file +7
-0+ description: + A container for devices not accessible via any data bus. Common use + cases include fixed and GPIO regulators, simple video panels and LED + or GPIO backlight devices. When not hot-pluggable, nodes such devices + are children of the root node. + + This node should not be present in the connector description in the
Quoted 8 lines
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml line 60
Quoted 5 lines
Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml new file +5
-0+ base device tree. It should be added by overlays along with a subnode + per device. + + type: object + additionalProperties: falseThe schema needs to work with the overlay applied too. 'true' is fine here.
Does additionalProperties apply to nodes as well? No properties are supposed to be added inside this node, only nodes, so I'm not sure what to do about additionalProperties.
Queued all other changes for v4.
Luca
Signature · 3 lines
Luca Ceresoli, Bootlin Embedded Linux and Kernel engineering https://bootlin.com
Luca Ceresoli
reviewerPATCH v3 2/7 drm/bridge: add bridge notifier to be notified of bridge addition and removal
In reply to Luca CeresoliFrom: Paul Kocialkowski <[email protected]>
In preparation for allowing bridges to be added to and removed from a DRM card without destroying the whole card, add a DRM bridge notifier. Notified events are addition and removal to/from the global bridge list.
---
Changes in v3: none
Changes in v2: none
--- drivers/gpu/drm/drm_bridge.c | 35 +++++++++++++++++++++++++++++++++++ include/drm/drm_bridge.h | 19 +++++++++++++++++++ 2 files changed, 54 insertions(+)
drivers/gpu/drm/drm_bridge.c +35 -0
@@ -25,6 +25,7 @@#include <linux/media-bus-format.h> #include <linux/module.h> #include <linux/mutex.h> +#include <linux/notifier.h> #include <drm/drm_atomic_state_helper.h> #include <drm/drm_bridge.h>
@@ -197,6 +198,36 @@static DEFINE_MUTEX (bridge_lock);static LIST_HEAD (bridge_list); +static BLOCKING_NOTIFIER_HEAD (bridge_notifier); + +/** +* drm_bridge_notifier_register - add a DRM bridge notifier +* @nb: the notifier block to be registered +* +* The notifier block will be notified of events defined in +* &drm_bridge_notifier_event +*/ +int drm_bridge_notifier_register (struct notifier_block *nb) +{ +return blocking_notifier_chain_register (&bridge_notifier, nb); +} +EXPORT_SYMBOL (drm_bridge_notifier_register); + +/** +* drm_bridge_notifier_unregister - remove a DRM bridge notifier +* @nb: the notifier block to be unregistered +*/ +int drm_bridge_notifier_unregister (struct notifier_block *nb) +{ +return blocking_notifier_chain_unregister (&bridge_notifier, nb); +} +EXPORT_SYMBOL (drm_bridge_notifier_unregister); + +static void drm_bridge_notifier_notify (unsigned long event, +struct drm_bridge *bridge) +{ +blocking_notifier_call_chain (&bridge_notifier, event, bridge); +}/** * drm_bridge_add - add the given bridge to the global bridge list
@@ -210,6 +241,8 @@ void drm_bridge_add(struct drm_bridge *bridge)mutex_lock (&bridge_lock);list_add_tail (&bridge->list, &bridge_list);mutex_unlock (&bridge_lock); + +drm_bridge_notifier_notify (DRM_BRIDGE_NOTIFY_ADD , bridge); }EXPORT_SYMBOL (drm_bridge_add);
@@ -243,6 +276,8 @@ EXPORT_SYMBOL(devm_drm_bridge_add);*/ void drm_bridge_remove (struct drm_bridge *bridge) { +drm_bridge_notifier_notify (DRM_BRIDGE_NOTIFY_REMOVE , bridge); +mutex_lock (&bridge_lock);list_del_init (&bridge->list);mutex_unlock (&bridge_lock);
include/drm/drm_bridge.h +19 -0
@@ -43,6 +43,22 @@ struct drm_panel;struct edid ;struct i2c_adapter ; +/** +* enum drm_bridge_notifier_event - DRM bridge events +*/ +enum drm_bridge_notifier_event { +/** +* @DRM_BRIDGE_NOTIFY_ADD: A bridge has just been added to the +* global bridge list. See drm_bridge_add(). +*/ +DRM_BRIDGE_NOTIFY_ADD , +/** +* @DRM_BRIDGE_NOTIFY_REMOVE: A bridge is about to be removed from +* the global bridge list. See drm_bridge_remove(). +*/ +DRM_BRIDGE_NOTIFY_REMOVE , +}; +/** * enum drm_bridge_attach_flags - Flags for &drm_bridge_funcs.attach */
@@ -862,6 +878,9 @@ drm_priv_to_bridge(struct drm_private_obj *priv) return container_of(priv, struct drm_bridge, base); } +int drm_bridge_notifier_register (struct notifier_block *nb); +int drm_bridge_notifier_unregister (struct notifier_block *nb); +void drm_bridge_add (struct drm_bridge *bridge);int devm_drm_bridge_add (struct device *dev,struct drm_bridge *bridge);void drm_bridge_remove (struct drm_bridge *bridge);
Signature · 1 line
2.34.1
Luca Ceresoli
reviewerPATCH v3 3/7 drm/encoder: add drm_encoder_cleanup_from()
In reply to Luca CeresoliSupporting hardware whose final part of the DRM pipeline can be physically removed requires the ability to detach all bridges from a given point to the end of the pipeline.
Introduce a variant of drm_encoder_cleanup() for this.
---
Changes in v3: none
Changed in v2:
- fix a typo in a comment
--- drivers/gpu/drm/drm_encoder.c | 21 +++++++++++++++++++++ include/drm/drm_encoder.h | 1 + 2 files changed, 22 insertions(+)
drivers/gpu/drm/drm_encoder.c +21 -0
@@ -207,6 +207,27 @@ void drm_encoder_cleanup(struct drm_encoder *encoder) }EXPORT_SYMBOL (drm_encoder_cleanup); +/** +* drm_encoder_cleanup_from - remove a given bridge and all the following +* @encoder: encoder whole list of bridges shall be pruned +* @bridge: first bridge to remove +* +* Removes from an encoder all the bridges starting with a given bridge +* and until the end of the chain. +* +* This should not be used in "normal" DRM pipelines. It is only useful for +* devices whose final part of the DRM chain can be physically removed and +* later reconnected (possibly with different hardware). +*/ +void drm_encoder_cleanup_from (struct drm_encoder *encoder,struct drm_bridge *bridge) +{ +struct drm_bridge *next; + +list_for_each_entry_safe_from (bridge, next, &encoder->bridge_chain, chain_node) +drm_bridge_detach (bridge); +} +EXPORT_SYMBOL (drm_encoder_cleanup_from); +static void drmm_encoder_alloc_release (struct drm_device *dev,void *ptr) {struct drm_encoder *encoder = ptr;
include/drm/drm_encoder.h +1 -0
@@ -320,6 +320,7 @@ static inline struct drm_encoder *drm_encoder_find(struct drm_device *dev, }void drm_encoder_cleanup (struct drm_encoder *encoder); +void drm_encoder_cleanup_from (struct drm_encoder *encoder,struct drm_bridge *bridge);/** * drm_for_each_encoder_mask - iterate over encoders specified by bitmask
Signature · 1 line
2.34.1
Luca Ceresoli
reviewerPATCH v3 4/7 drm/bridge: hotplug-bridge: add driver to support hot-pluggable DSI bridges
In reply to Luca CeresoliThis driver implements the point of a DRM pipeline where a connector allows removal of all the following bridges up to the panel.
The DRM subsystem currently allows hotplug of the monitor but not preceding components. However there are embedded devices where the "tail" of the DRM pipeline, including one or more bridges, can be physically removed:
.------------------------.
| DISPLAY CONTROLLER |
| .---------. .------. |
| | ENCODER |<--| CRTC | |
| '---------' '------' |
'------|-----------------'
|
| HOTPLUG
V CONNECTOR
.---------. .--. .-. .---------. .-------.
| 0 to N | | _| _| | | 1 to N | | |
| BRIDGES |--DSI-->||_ |_ |--DSI-->| BRIDGES |--LVDS-->| PANEL |
| | | | | | | | | |
'---------' '--' '-' '---------' '-------'[--- fixed components --] [----------- removable add-on -----------]
This driver supports such a device, where the final segment of a MIPI DSI bus, including one or more bridges, can be physically disconnected and reconnected at runtime, possibly with a different model.
The add-on supported by this driver has a MIPI DSI bus traversing the hotplug connector and a DSI to LVDS bridge and an LVDS panel on the add-on. Hovever it is designed to be as far as possible generic and extendable to other busses that have no native hotplug and model ID discovery.
This driver does not itself add and remove the bridges or panel on the
add-on: this needs to be done by other means, and in this implementation it is done by device tree overlay runtime insertion and removal. The hotplug-bridge gets notified of hot-plugging by the DRM bridge notifier callbacks after they get added or before they get removed.
The hotplug-bridge role is to implement the "hot-pluggable connector" in the bridge chain. In this position, what the hotplug-bridge should ideally do is:
* communicate with the previous component (bridge or encoder) so that it believes it always has a connected bridge following it and the DRM card is always present
* be notified of the addition and removal of the following bridge and attach/detach to/from it
* communicate with the following bridge so that it will attach and detach using the normal procedure (as if the entire pipeline were being created or destroyed, not only the tail)
* instantiate two DRM connectors (similarly to what the DisplayPort MST code does):
- a DSI connector representing the video lines of the hotplug connector; the status is always "disconnected" (no panel is ever attached directly to it)
- an LSVD connector representing the classic connection to the panel; this gets added/removed whenever the add-on gets connected/disconnected; the status is always "connected" as the panel is always connected when the add-on is
However some aspects make it a bit more complex than that. Most notably:
* the next bridge can be probed and removed at any moment and all probing sequences need to be handled
* the DSI host/device registration process, which adds to the DRM bridge attach process, makes the initial card registration tricky
* the need to register and deregister the following bridges at runtime without tearing down the whole DRM card prevents using some of the functions that are normally recommended
* the automatic mechanism to call the appropriate .get_modes operation (typically provided by the panel bridge) cannot work as the panel can disappear and reappear as a different model, so an ad-hoc lookup is needed
The code handling these and other tricky aspects is accurately documented by comments in the code.
---
Changed in v3:
- dynamically add/remove the LVDS connector on hot(un)plug
- take the firmware node normally via dev->of_node instead of using device_set_node(); this makes code more self-contained and generic
- minor rewordings and cleanups
Changed in v2:
- change to be a platform device instantiated from the connector driver instead of a self-standing OF driver
- add missing error handling for devm_drm_bridge_add()
- various cleanups and style improvements
--- MAINTAINERS | 5 + drivers/gpu/drm/bridge/Kconfig | 15 + drivers/gpu/drm/bridge/Makefile | 1 + drivers/gpu/drm/bridge/hotplug-bridge.c | 665 ++++++++++++++++++++++++++++++++ 4 files changed, 686 insertions(+)
MAINTAINERS +5 -0
@@ -7004,6 +7004,11 @@ T: git https://gitlab.freedesktop.org/drm/misc/kernel.git F: Documentation/devicetree/bindings/display/panel/himax,hx8394.yaml F: drivers/gpu/drm/panel/panel-himax-hx8394.c +DRM DRIVER FOR HOTPLUG VIDEO CONNECTOR BRIDGE +M: Luca Ceresoli <[email protected]> +S: Maintained +F: drivers/gpu/drm/bridge/hotplug-bridge.c + DRM DRIVER FOR HX8357D PANELS S: Orphan T: git https://gitlab.freedesktop.org/drm/misc/kernel.git
drivers/gpu/drm/bridge/Kconfig +15 -0
@@ -90,6 +90,21 @@ config DRM_FSL_LDB help Support for i.MX8MP DPI-to-LVDS on-SoC encoder. +config DRM_HOTPLUG_BRIDGE + tristate "Hotplug DRM bridge support" + depends on OF + select DRM_PANEL_BRIDGE + select DRM_MIPI_DSI + select DRM_KMS_HELPER + help + Driver for a DRM bridge representing a physical connector that + splits a DRM pipeline into a fixed part and a physically + removable part. The fixed part includes up to the encoder and + zero or more bridges. The removable part includes any following + bridges up to the connector and panel and can be physically + removed and connected at runtime, possibly with different + components. + config DRM_ITE_IT6505 tristate "ITE IT6505 DisplayPort bridge" depends on OF
drivers/gpu/drm/bridge/Makefile +1 -0
@@ -6,6 +6,7 @@ obj-$(CONFIG_DRM_CHRONTEL_CH7033) += chrontel-ch7033.o obj-$(CONFIG_DRM_CROS_EC_ANX7688) += cros-ec-anx7688.o obj-$(CONFIG_DRM_DISPLAY_CONNECTOR) += display-connector.o obj-$(CONFIG_DRM_FSL_LDB) += fsl-ldb.o +obj-$(CONFIG_DRM_HOTPLUG_BRIDGE) += hotplug-bridge.o obj-$(CONFIG_DRM_ITE_IT6505) += ite-it6505.o obj-$(CONFIG_DRM_LONTIUM_LT8912B) += lontium-lt8912b.o obj-$(CONFIG_DRM_LONTIUM_LT9211) += lontium-lt9211.o
drivers/gpu/drm/bridge/hotplug-bridge.c new file +665 -0
665 lines not on this page · raw message
Signature · 1 line
2.34.1
Luca Ceresoli
reviewerPATCH v3 5/7 i2c: i2c-core-of: follow i2c-parent phandle to probe devices from added nodes
In reply to Luca CeresoliWhen device tree nodes are added, the I2C core tries to probe client devices based on the classic DT structure:
i2c@abcd0000 {
some-client@42 { compatible = "xyz,blah"; ... };
};However for hotplug connectors described via device tree overlays there is additional level of indirection, which is needed to decouple the overlay and the base tree:
--- base device tree ---
i2c1: i2c@abcd0000 { compatible = "xyz,i2c-ctrl"; ... };
i2c5: i2c@cafe0000 { compatible = "xyz,i2c-ctrl"; ... };
connector {
i2c-ctrl {
i2c-parent = <&i2c1>;
#address-cells = <1>;
#size-cells = <0>;
}; i2c-sensors {
i2c-parent = <&i2c5>;
#address-cells = <1>;
#size-cells = <0>;
};};
--- device tree overlay ---
...
// This node will overlay on the i2c-ctrl node of the base tree
i2c-ctrl {
eeprom@50 { compatible = "atmel,24c64"; ... };
};
...--- resulting device tree ---
i2c1: i2c@abcd0000 { compatible = "xyz,i2c-ctrl"; ... };
i2c5: i2c@cafe0000 { compatible = "xyz,i2c-ctrl"; ... };
connector {
i2c-ctrl {
i2c-parent = <&i2c1>;
#address-cells = <1>;
#size-cells = <0>; eeprom@50 { ompatible = "atmel,24c64"; ... };
};
i2c-sensors {
i2c-parent = <&i2c5>;
#address-cells = <1>;
#size-cells = <0>;
};};
Here i2c-ctrl (same goes for i2c-sensors) represent the part of I2C bus that is on the hot-pluggable add-on. On hot-plugging it will physically connect to the I2C adapter on the base board. Let's call the 'i2c-ctrl' node an "extension node".
In order to decouple the overlay from the base tree, the I2C adapter (i2c@abcd0000) and the extension node (i2c-ctrl) are separate nodes. Rightfully, only the former will probe into an I2C adapter, and it will do that perhaps during boot, long before overlay insertion.
The extension node won't probe into an I2C adapter or any other device or bus, so its subnodes ('eeprom@50') won't be interpreted as I2C clients by current I2C core code. However it has an 'i2c-parent' phandle to point to the corresponding I2C adapter node. This tells those nodes are I2C clients of the adapter in that other node.
Extend the i2c-core-of code to look for the adapter via the 'i2c-parent' phandle when the regular adapter lookup does not find one. This allows all clients to be probed: both those on the base board (described in the base device tree) and those on the add-on and described by an overlay.
---
Note: while this patch works for normal hotplug and unplug, it has some weaknesses too, due to the implementation being in a OF change notifier. Two cases come to mind:
1. In the base device tree there must be _no_ nodes under the "extension node" (i2c-ctrl), or they won't be picked up as they are not dynamically added.
2. In case the I2C adapter is unbound and rebound, or it probes after overlay insertion, it will miss the OF notifier events and so it won't find the devices in the extension node.
The first case is not a limiting factor: fixed I2C devices should just stay under the good old I2C adapter node.
The second case is a limiting factor, even though not happening in "normal" use cases. I cannot see any solution without making the adapter aware of the "bus extensions" it has, so on its probe it can always go look for any devices there. Taking into account the case of multiple connectors each having an extension of the same bus, this may look as follows in device tree:
--- base device tree ---
i2c1: i2c@abcd0000 {
compatible = "xyz,i2c-ctrl"; ...
i2c-bus-extensions = <&i2c_ctrl_conn0, &i2c_ctrl_conn1>;
};
connector@0 {
i2c_ctrl_conn0: i2c-ctrl {
i2c-parent = <&i2c1>;
#address-cells = <1>;
#size-cells = <0>;
};
};
connector@1 {
i2c_ctrl_conn1: i2c-ctrl {
i2c-parent = <&i2c1>;
#address-cells = <1>;
#size-cells = <0>;
};
};I'd love to have some feedback and opinions about the basic idea before digging into the details of this additional step.
---
This patch first appeared in v3.
--- drivers/i2c/i2c-core-of.c | 9 +++++++++ 1 file changed, 9 insertions(+)
drivers/i2c/i2c-core-of.c +9 -0
@@ -170,6 +170,15 @@ static int of_i2c_notify(struct notifier_block *nb, unsigned long action,switch (of_reconfig_get_state_change (action, rd)) {case OF_RECONFIG_CHANGE_ADD : adap =of_find_i2c_adapter_by_node (rd->dn->parent); +if (adap ==NULL ) { +struct device_node *i2c_bus; + + i2c_bus =of_parse_phandle (rd->dn->parent,"i2c-parent" ,0 ); +if (i2c_bus) { + adap =of_find_i2c_adapter_by_node (i2c_bus); +of_node_put (i2c_bus); + } + }if (adap ==NULL )return NOTIFY_OK ;/* not for us */
Signature · 1 line
2.34.1
Luca Ceresoli
reviewerPATCH v3 6/7 misc: add ge-addon-connector driver
In reply to Luca CeresoliAdd a driver to support the runtime hot-pluggable add-on connector on the GE SUNH device. This connector allows connecting and disconnecting an add-on to/from the main device to augment its features. Connection and disconnection can happen at runtime at any moment without notice.
Different add-on models can be connected, and each has an EEPROM with a model identifier at a fixed address.
The add-on hardware is added and removed using device tree overlay loading and unloading.
---
Changed in v3:
- update to the new overlay representation that now does not include the target node; instead the target node is the connector node itself and is now passed by the connector driver to of_overlay_fdt_apply(), so the overlay is now decoupled from the base device tree
- update to explicitely probe devices not reachable by the CPU on any physical bus (which probe as platform devices) which are now inside a 'nobus-devices' subnode of the connector node
- change how the DRM bridge is populated to use the new device tree representation, having the video ports inside the 'dsi' node
**NOTE** this specific change opens up a question about the
.of_node_reused flag: setting it to true might be wrong now as the bridge will be handed the 'dsi' subnode of the connector node; however not setting it to true prevents the hotplug bridge module autoloading due to the alias string changing from "platform:hotplug-dsi-bridge" to "of:NdsiT(null)".
- remove dev_info() and uninformative dev_dbg() calls
- Kconfig: use 'depends on' instead of 'select'
- Kconfig: improve help text and add module name
This patch first appeared in v2.
--- MAINTAINERS | 1 + drivers/misc/Kconfig | 18 ++ drivers/misc/Makefile | 1 + drivers/misc/ge-sunh-connector.c | 523 +++++++++++++++++++++++++++++++++++++++ 4 files changed, 543 insertions(+)
MAINTAINERS +1 -0
@@ -10262,6 +10262,7 @@ F: drivers/iio/pressure/mprls0025pa* HOTPLUG CONNECTOR FOR GE SUNH ADDONS M: Luca Ceresoli <[email protected]> S: Maintained +F: drivers/misc/ge-sunh-connector.c F: Documentation/devicetree/bindings/connector/ge,sunh-addon-connector.yaml HP BIOSCFG DRIVER
drivers/misc/Kconfig +18 -0
@@ -600,6 +600,24 @@ config MARVELL_CN10K_DPI To compile this driver as a module, choose M here: the module will be called mrvl_cn10k_dpi. +config GE_SUNH_CONNECTOR + tristate "GE SUNH hotplug add-on connector" + depends on OF_OVERLAY + depends on NVMEM + depends on DRM_HOTPLUG_BRIDGE + select FW_LOADER + help + Driver for the runtime hot-pluggable add-on connector on the GE + SUNH device. This connector allows connecting an add-on to the + main device to augment its features, and to later disconnect + it. Connection and disconnection can be done at runtime at any + moment without notice. Different add-on models can be connected, + and each has an EEPROM with a model identifier at a fixed + address. + + To compile this driver as a module, choose M here. + The module will be called ge-sunh-connector. + source "drivers/misc/c2port/Kconfig" source "drivers/misc/eeprom/Kconfig" source "drivers/misc/cb710/Kconfig"
drivers/misc/Makefile +1 -0
@@ -70,4 +70,5 @@ obj-$(CONFIG_TPS6594_ESM) += tps6594-esm.o obj-$(CONFIG_TPS6594_PFSM) += tps6594-pfsm.o obj-$(CONFIG_NSM) += nsm.o obj-$(CONFIG_MARVELL_CN10K_DPI) += mrvl_cn10k_dpi.o +obj-$(CONFIG_GE_SUNH_CONNECTOR) += ge-sunh-connector.o obj-y += keba/
drivers/misc/ge-sunh-connector.c new file +523 -0
523 lines not on this page · raw message
Signature · 1 line
2.34.1
Luca Ceresoli
reviewerPATCH v3 7/7 driver core: do not unblock consumers any drivers found
In reply to Luca CeresoliQuick summary:
I have investigated a problem for a long time, I have a pretty good understanding, I tried various fixes but none except this is working. The goal of this patch is to discuss the problem to converge to the best solution.
---------------------------------
Symptoms
---------------------------------
The problem appeared while testing the v3 addon connector driver that is part of this seres, and which is based on device tree overlays.
Note the symptom happens often, but not always. Changes to logging is a typical way to make it appear/disappear, so it appears as time sensitive.
The relevant DT overlay snippet is:
/ {
fragment@0 {
target-path = "";
__overlay__ {
nobus-devices {
// nodes in here are populated as platform devices
reg_addon_3v3_lcd: regulator-addon-3v3-lcd {
compatible = "regulator-fixed";
regulator-name = "3V3_LCD_ADDON";
gpios = <...>;
};
addon_panel_dsi_lvds: panel-dsi-lvds {
compatible = "...";
power-supply = <®_addon_3v3_lcd>;
};
};
};
};
};So the regulator is a supplier to the panel. Nothing special here, except we are in an overlay.
The overlay gets applied and all devices work correctly. Troubles start appearing in the form of two messages on overlay removal, in this order:
* WARNING: CPU: 1 PID: 189 at drivers/regulator/core.c:5856 regulator_unregister+0x1ec/0x208
This is issued during removal of the 3V3_LCD_ADDON regulator because rdev->open_count is 1, while it should be 0. This is because the panel still hasn't closed the regulator.
* Unable to handle kernel NULL pointer dereference at virtual address 0000000000000000
Call trace:
regulator_lock_recursive+0x5c/0x200
regulator_lock_dependent+0xc0/0x140
regulator_enable+0x44/0x98
panel_simple_resume+0x38/0x108 [panel_simple]
pm_generic_runtime_resume+0x34/0x58
__rpm_callback+0x50/0x1f0
rpm_callback+0x70/0x88
rpm_resume+0x49c/0x678
__pm_runtime_resume+0x54/0xa0
device_release_driver_internal+0xd4/0x240
device_release_driver+0x20/0x38
bus_remove_device+0xd4/0x120
device_del+0x154/0x388This happens while the panel driver is being removed and the devm infra tries to close the regulator which is already gone.
Both errors have the same origin: the regulator driver is removed before the panel driver.
---------------------------------
Problem analysis
---------------------------------
My analysis showed that the problem originates from devlink manipulation during overlay insertion, but shows its effects on removal. This is the sequence of events:
* During overlay insertion:
1. the devlink code creates a devlink (A) with:
supplier = regulator-addon-3v3-lcd
consumer = panel-dsi-lvds
flags = DL_FLAG_INFERRED (+ possibly others) because it has
been inferred from firmware data
2. soon after, devlink A is relaxed and then dropped
- does not happen always, based on timing
- see below for details
3. the regulator-addon-3v3-lcd regulator gets probed as a platform device
- the probe function for regulator-addon-3v3-lcd is reg_fixed_voltage_probe(), which calls devm_regulator_register() to register a single new regulator class device for the voltage output
- regulator_register() does, among others:
- instantiate a new regulator class device (3V3_LCD_ADDON), with parent = regulator-addon-3v3-lcd
- adds a devlink (B) with:
supplier = 3V3_LCD_ADDON
consumer = panel-dsi-lvds
At this point we have these devices and devlinks:
.---------------------------.
| regulator-addon-3v3-lcd |
| regulator platform device | supplier consumer
| (struct device) |<--------- devlink A -------------.
'---------------------------' (inferred) |
^ V
| .-----------------.
| parent | panel-dsi-lvds |
| | (struct device) |
| '-----------------'
.---------------------------. ^
| 3V3_LCD_ADDON | supplier consumer |
| regulator class device |<--------- devlink B -------------'
| (struct device) | (created by
'---------------------------' regulator core)Depending on whether step 2 happens or not, devlink A will be still present or not during overlay removal.
When step 2 happens and devlink A gets dropped (which happens to me almost always), the removal code calls:
-> device_release_driver(dev = regulator-addon-3v3-lcd)
-> device_release_driver_internal()
-> __device_release_driver()
-> if (device_links_busy()) // see below
{
device_links_unbind_consumers(dev = regulator-addon-3v3-lcd)
-> for each consumer for which 'dev' is a supplier:
{
device_release_driver_internal(dev = panel-dsi-lvds)
}
}The logic is pretty clear: before removing a device that is a supplier to other devices (regulator-addon-3v3-lcd), use devlink to find all consumers (panel-dsi-lvds) and remove them first, recursively.
However in case devlink A had been initially dropped, there is no devlink between the two devices. The regulator removal will just proceed, and the regulator device gets removed before its consumer.
Note devlink B is not at all within this removal phase. device_links_busy() looks at the platform device (regulator-addon-3v3-lcd), and it has no way to know about the class device (3V3_LCD_ADDON).
Assuming the whole device_links_busy() / device_links_unbind_consumers() logic is correct, let's move to why devlink A gets dropped.
---------------------------------
Why the devlink is dropped
---------------------------------
It all starts in the device_add() for regulator-addon-3v3-lcd:
/*
* If all driver registration is done and a newly added device doesn't
* match with any driver, don't block its consumers from probing in
* case the consumer device is able to operate without this supplier. */
if (dev->fwnode && fw_devlink_drv_reg_done && !dev->can_match)
fw_devlink_unblock_consumers(dev);
The three conditions in the if() mean:
1. this device comes from firmware -> always true in my case (device tree)
2. this global flag is set via the deferred_probe_timeout_work as soon as for 10 consecutive seconds there is no new driver being probed; it is never cleared
3. no driver has been matched with this device so far (IOW the probe function of the driver for this device has never been called, regardless of the return value)
If all condtions apply, fw_devlink_unblock_consumers() will (after some checks) call fw_devlink_relax_link() on every link to consumers and "relax" it. Relaxing means setting link flags to DL_FLAG_MANAGED | FW_DEVLINK_FLAGS_PERMISSIVE. Soon later, device_links_driver_bound() will take devlinks with these flags and drop them.
I was unable to understand in full detail the flag manipulation logic happening in the devlink code. However I think the high-level logic here can be expressed as: if a devlink was inferred from firmware and its supplier device did not probe after a while (*) because no potential driver was found, then maybe that devlink was wrong or it is enforcing a supplier that is optional for the consumer: let's drop the link and see whether the (formerly devlink consumer) device can now probe.
(*) "after a while" is implemented by the fw_devlink_drv_reg_done flag,
which typically gets set way less than a minute after boot
Basically fw_devlink_drv_reg_done flag splits the probing in two phases. In phase 1 we try to probe all inferred suppliers before probing consumers. Then we set fw_devlink_drv_reg_done and relax+drop the "dangling" inferred devlinks. Then in phase 2 we try to probe without inferred devlinks. This is to see if we can probe more devices due to incorrectly inferred devlinks or missing drivers for optional suppliers.
Overlays however can be loaded at any time, even a long time after booting. This is totally normal when used for a hotplug connector, where the devices get physically connected by the user. This implies the fw_devlink_drv_reg_done flag is found already set when probing overlay devices. And so, conditions 1 and 2 above are always set in the overlay case.
So we are left with condition 3 only. Again I haven't done a full analysis here, but it is perfectly fine that a driver is not immediately present when adding a new device. It can just have not yet been matched, possibly because a driver module is in process of being loaded from storage.
I think there is a race here: on one side the driver becoming available and matched and the device to probe, on the other side the fw_devlink_unblock_consumers() logic to relax and drop inferred devlinks. If the device probes first, the link won't be dropped.
---------------------------------
Same problem without DT overlays?
---------------------------------
Based on the above, I suspect the exact same problem exists even without any overlay usage. Indeed, the conditions might exist in other corner cases. One example is a device whose driver is a module and is not loaded during boot: e.g. it is not on the root filesystem, it is being developed and the programmer sends the driver via SSH to a tmpfs to load and test it.
As said, this is a matter of corner cases, but still possible.
Note that no problem should happen to natively removable devices such as USB, because condition 1 defuses the whole if() above for devices not described in firmware.
---------------------------------
Fixes I have tried (not working)
---------------------------------
I tried a couple approaches based on devlink to fix this issue.
One was augmenting the regulator core code to add a new devlink between the regulator platform device (regulator-addon-3v3-lcd) and the regulator class device (3V3_LCD_ADDON), to create a chain for device_links_busy() to follow. The devlink is created and persists until removal time. However it does not enforce the correct ordering: device_links_busy() ignores it because the link status is always "dormant". The reason appears to be that the "regulator output device" is a struct device but never has a driver. Recently Saravana pointed out that:
Quoted 1 line
device links don't work correctly for "class" type devices
(https://lore.kernel.org/all/CAGETcx-ioF=jTbyQMeD2fsYKz8q5vw_TWYWS9m8H5=pCo5KFYA@mail.gmail.com/)
which is possibly related.
I tried a variant: change the devlink already created by _regulator_get() to use the regulator platform device (regulator-addon-3v3-lcd) instead of the regulator class device (3V3_LCD_ADDON) as the supplier. That would make devlink B have the same endpoints as devlink A. However this did not work due to the link state staying "not tracked" and thus again being ignored by device_links_busy(). I haven't managed to find out the flag manipulations that would make it work.
---------------------------------
Conclusions
---------------------------------
The current logic is clearly OK for "typical" current use cases (not counting corner cases), but unsuitable for hotplugged devices described by firmware.
The question is: do we have an underlying assumption that was valid so far but is wrong when overlays are added?
One possible answer is: dropping inferred devlinks is wrong. Generally speaking, inferred devlinks are sometimes useless but don't hurt, so there is no need to drop them. This is what this patch changes.
However I realize there is a use case for dropping inferred devlink:
optional suppliers that prevent consumer probing until they are dropped. Indeed, inferring devlinks from firmware data can create links that prevent some device to probe. For this reason my first attempts have been to add or change the devlinks that subsystem code creates.
So a more sophisticated idea is that after phase 1 we try to probe all not-probed-yet consumers ignoring the relaxed devlinks, instead of removing them. This would allow the same amount of devices to be probed using the same amount of optional suppliers, but leaving the inferred devlinks in place because they might be useful later on.
And then of course there are the above solutions I failed to get working, which might be the right way but need some directions for me to have them working.
I am very open to more answers and suggestions.
Best regards,
Luca
---
This patch first appeared in v3.
--- drivers/base/core.c | 21 --------------------- 1 file changed, 21 deletions(-)
drivers/base/core.c +0 -21
@@ -1875,19 +1875,6 @@ void __init wait_for_init_devices_probe(void) fw_devlink_best_effort = false; } -static void fw_devlink_unblock_consumers (struct device *dev) -{ -struct device_link *link; - -if (!fw_devlink_flags ||fw_devlink_is_permissive ()) -return ; - -device_links_write_lock (); -list_for_each_entry (link, &dev->links.consumers, s_node) -fw_devlink_relax_link (link); -device_links_write_unlock (); -} -#define get_dev_from_fwnode (fwnode) get_device((fwnode)->dev)static bool fwnode_init_without_drv (struct fwnode_handle *fwnode)
@@ -3679,14 +3666,6 @@ int device_add(struct device *dev)bus_probe_device (dev); -/* -* If all driver registration is done and a newly added device doesn't -* match with any driver, don't block its consumers from probing in -* case the consumer device is able to operate without this supplier. -*/ -if (dev->fwnode && fw_devlink_drv_reg_done && !dev->can_match) -fw_devlink_unblock_consumers (dev); -if (parent)klist_add_tail (&dev->p->knode_parent, &parent->p->klist_children);
Signature · 1 line
2.34.1
Luca Ceresoli
reviewerHello Rob,
On Fri, 09 Aug 2024 17:34:48 +0200
Luca Ceresoli <[email protected]> wrote:
...
Quoted 6 lines
However a few new rough edges emerged that are not yet solved in this v3. Discussion would help in finding the right direction:
* Describing the NVMEM cell addition still requires adding two properties to a node in the base tree. Not sure the current NVMEM cell bindings allow to do better.
Do you have any thoughts about how to describe the NVMEM cell in DT without adding properties?
As of now...
Quoted 6 lines
2: the "base" overlay
The "base" overlay describes the common components that are required to read the model ID. These are identical for all add-on models, thus only one "base" overlay is needed:
Quoted 10 lines
/dts-v1/;
/plugin/;
/ {
fragment@0 {
target-path = "";
__overlay__ {
nvmem-cells = <&addon_id>;
nvmem-cell-names = "id";...the nvmem-cell* properties are the only ones that for lack of a good solution get added to a node in the base tree, causing the memory leak issue to be still present.
Luca
Signature · 3 lines
Luca Ceresoli, Bootlin Embedded Linux and Kernel engineering https://bootlin.com