Skip to contentSkip to conversation

Support ACPI and SETAASA device discovery

PATCH Applied Akhil R 3 participants 28 messages Jul 28 – Jul 30

Subsystems (5)
lore
Share & tools
lore

Akhil R

author
#0 lore raw

v7 cover Support ACPI and SETAASA device discovery

This patch series adds SETAASA device discovery to the I3C subsystem, enabling support for SPD5118 temperature sensors found on DDR5 memory modules. The changes also add ACPI support for all existing DAA methods like SETDASA, SETNEWDA as well as I2C devices on I3C bus.

SPD5118 and similar devices on DDR5 memory modules differ from typical I3C devices in their initialization. They use SETAASA broadcast CCC instead of ENTDAA for address assignment, and per JEDEC specification, are not required to have a Provisioned ID or implement standard device information CCC commands (GETPID, GETDCR, GETBCR).

The series enables describing all I3C and I2C devices on both Device Tree and the ACPI table, using unified device property APIs throughout the I3C core and the Synopsys DesignWare I3C master driver.

Please note that the series modifies drivers across multiple subsystems, like Device Tree bindings, ACPI, I3C and HWMON.

v6->v7:
* Reject I3C devices with a zero manufacturer ID by validating with I3C_PID_MANUF_ID() instead of the whole PID.
* Keep the static address reserved when a SETAASA device cannot be added, so that it is not assigned to another device during DAA.
* Do not issue RSTDAA when a SETAASA device fails to attach, as no address assignment CCC was sent to it.
* Drop the redundant core clock check in dw_i3c_master_get_core_rate() as probe already fails when the clock is missing without the ACPI skip quirk.

v5->v6:
* Do not take an extra fwnode reference when assigning the boardinfo fwnode to a newly registered I3C device.
* Issue SETAASA after SETDASA pre-assignment and before ENTDAA.
* Prefer SETDASA over SETAASA when a device advertises both methods and an explicit dynamic address (assigned-address) is provided.
* Reject PID 0 for non-SETAASA I3C boardinfo entries.
* Drop OF/ACPI uevent modalias emission; keep OF/ACPI matching for SETAASA devices only.
* Resolve DesignWare quirks before core clock acquisition and fail probe early when the clock is missing without the ACPI skip quirk.

v4->v5:
* Free ACPI I2C boardinfo when an ACPI child without an I2C resource is ignored.
* Issue RSTDAA if SETAASA device attach fails after the static address is used as the dynamic address.
* Emit OF and ACPI modaliases for firmware-matched I3C devices.
* Make the DesignWare core clock optional and use "clock-frequency" as the fallback when core_clk is absent.
* Keep DesignWare pclk, reset, and match-data handling on their existing optional paths.

v3->v4:
* Clarify mipi-i3c-static-method bit semantics and assigned-address
* Add I3C_ADDR_METHOD_VENDOR
* Fix fwnode reference handling while converting child property parsing to use unified firmware-node APIs.
* Align ACPI child enumeration with the I2C core for multiple I2cSerialBus resources, ignore ACPI child entries without an I2C resource, and populate I2C modalias information from ACPI.
* Update SETAASA handling to use the static address as the dynamic address, skip device-info retrieval for SETAASA devices, and tolerate M2 for SETHID/SETAASA similarly to ENTDAA.
* Reorder DesignWare I3C clock/reset to include optional clock in the ACPI skip clock/reset quirk.
* Add prints for missing ACPI clock-frequency and SPD5118 I3C device type read failures.
* Fix grammar in comments and commit messages.

v2->v3:
* Fix maximum value and indent bit list for mipi-i3c-static-method.
* Move I3C_ADDR_METHOD_* macros to dt-bindings header.
* Drop ACPICA commit IDs, keep only the Link: tags.
* Revert the change which proceeds to register other devices if SETAASA is not supported so that it aligns with the rest of the driver and to avoid the issues pointed by Sashiko.
* Rework multiple commit messages.

v1->v2:
* Added patch to remove 16-bit addressing support for SPD5118
* Guard ACPI calls with #ifdef CONFIG_ACPI
* Remove CONFIG_OF guard for of_alias_get_highest_id()
* Mask mipi-i3c-static-method in the driver to select only valid values.
* Proceed to register other devices if SETAASA is not supported.
* Update commit message and links in the description of multiple commits.

Akhil R (12):
dt-bindings: i3c: Add mipi-i3c-static-method to support SETAASA
i3c: master: Use unified device property interface
i3c: master: Support ACPI enumeration of child devices
i3c: master: Add support for devices using SETAASA
i3c: master: Add support for devices without PID
i3c: master: match I3C device through DT and ACPI
i3c: dw-i3c-master: Add SETAASA as supported CCC
i3c: dw-i3c-master: Add ACPI core clock frequency quirk
i3c: dw-i3c-master: Add ACPI ID for Tegra410
hwmon: spd5118: Remove 16-bit addressing
hwmon: spd5118: Add I3C support
arm64: defconfig: Enable I3C and SPD5118 hwmon

 .../devicetree/bindings/i3c/i3c.yaml          |  36 +-
 arch/arm64/configs/defconfig                  |   3 +
 drivers/hwmon/Kconfig                         |   9 +-
 drivers/hwmon/spd5118.c                       | 119 +++--
 drivers/i3c/master.c                          | 406 +++++++++++++++---
 drivers/i3c/master/dw-i3c-master.c            |  54 ++-
 include/dt-bindings/i3c/i3c.h                 |   4 +
 include/linux/i3c/ccc.h                       |   1 +
 include/linux/i3c/master.h                    |  20 +-
 9 files changed, 507 insertions(+), 145 deletions(-)

base-commit: 6eb8711ece2ce27e52e327a5b7a628ed39b97f45

Signature · 1 line
2.43.0
author
#1 lore raw

PATCH v7 1/12 dt-bindings: i3c: Add mipi-i3c-static-method to support SETAASA

Add the 'mipi-i3c-static-method' property mentioned in the MIPI I3C Discovery and Configuration Specification [1] to specify which discovery method an I3C device supports during bus initialization. The property is a bitmap, where a bit value of 1 indicates support for that method, and 0 indicates lack of support.

Bit 0: SETDASA CCC (Direct)
Bit 1: SETAASA CCC (Broadcast)
Bit 2: Other CCC (vendor / standards extension)
All other bits are reserved.

It is specifically needed when an I3C device requires SETAASA for the address assignment. SETDASA will be supported by default if this property is absent, which means for now the property just serves as a flag to enable SETAASA, but keep the property as a bitmap to align with the specifications.

[1] https://www.mipi.org/mipi-disco-for-i3c-download

---
 .../devicetree/bindings/i3c/i3c.yaml          | 36 ++++++++++++++++---
 include/dt-bindings/i3c/i3c.h                 |  4 +++
 2 files changed, 35 insertions(+), 5 deletions(-)
Documentation/devicetree/bindings/i3c/i3c.yaml +31 -5
index e25fa72fd785..5603f2e7807d 100644
@@ -31,10 +31,12 @@ properties:
       described in the device tree, which in turn means we have to describe
       I3C devices.
 
-      Another use case for describing an I3C device in the device tree is when
-      this I3C device has a static I2C address and we want to assign it a
-      specific I3C dynamic address before the DAA takes place (so that other
-      devices on the bus can't take this dynamic address).
+      Other use-cases for describing an I3C device in the device tree are:
+      - When the I3C device has a static I2C address and we want to assign
+        it a specific I3C dynamic address before the DAA takes place (so
+        that other devices on the bus can't take this dynamic address).
+      - When the I3C device requires SETAASA for its discovery and uses a
+        pre-defined static address.
 
   "#size-cells":
     const: 0
@@ -145,7 +147,31 @@ patternProperties:
           Dynamic address to be assigned to this device. In case static address is
           present (first cell of the reg property != 0), this address is assigned
           through SETDASA. If static address is not present, this address is assigned
-          through SETNEWDA after assigning a temporary address via ENTDAA.
+          through SETNEWDA after assigning a temporary address via ENTDAA. If
+          SETAASA is used, this property is not used, and the static address itself
+          becomes the dynamic address.
+
+      mipi-i3c-static-method:
+        $ref: /schemas/types.yaml#/definitions/uint32
+        minimum: 0x1
+        maximum: 0x7
+        default: 1
+        description: |
+          Bitmap describing which methods of Dynamic Address Assignment from a
+          static address are supported by this I3C Target. For each defined bit
+          position, a set bit indicates support for that method and a cleared
+          bit indicates lack of support.
+
+            Bit 0: SETDASA CCC (Direct)
+            Bit 1: SETAASA CCC (Broadcast)
+            Bit 2: Other CCC (vendor / standards extension)
+            All other bits are reserved.
+
+          This property follows the MIPI I3C specification. The primary use
+          of this property is to indicate support for SETAASA, i.e Bit 1, but
+          will allow other values mentioned in the specification so that it
+          mirrors the specification. SETDASA will remain as the default method
+          even if this property is not present.
 
     required:
       - reg
include/dt-bindings/i3c/i3c.h +4 -0
index 373439218bba..78b8c634aad8 100644
@@ -13,4 +13,8 @@
 #define I2C_NO_FILTER_HIGH_FREQUENCY    (1 << 5)
 #define I2C_NO_FILTER_LOW_FREQUENCY     (2 << 5)
 
+#define I3C_ADDR_METHOD_SETDASA     (1 << 0)
+#define I3C_ADDR_METHOD_SETAASA     (1 << 1)
+#define I3C_ADDR_METHOD_VENDOR      (1 << 2)
+
 #endif
Signature · 1 line
2.43.0
#2 lore raw

Re: [PATCH v7 01/12] dt-bindings: i3c: Add mipi-i3c-static-method to support SETAASA

Sashiko has reviewed this patch and found no issues. It looks great!
Quoted 17 lines

Add the 'mipi-i3c-static-method' property mentioned in the MIPI I3C Discovery and Configuration Specification [1] to specify which discovery method an I3C device supports during bus initialization. The property is a bitmap, where a bit value of 1 indicates support for that method, and 0 indicates lack of support.

Bit 0: SETDASA CCC (Direct)
Bit 1: SETAASA CCC (Broadcast)
Bit 2: Other CCC (vendor / standards extension)
All other bits are reserved.

It is specifically needed when an I3C device requires SETAASA for the address assignment. SETDASA will be supported by default if this property is absent, which means for now the property just serves as a flag to enable SETAASA, but keep the property as a bitmap to align with the specifications.

Quoted 5 lines

[1] https://www.mipi.org/mipi-disco-for-i3c-download

Sashiko has reviewed this patch and found no issues. It looks great!

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=1
author
#3 lore raw

PATCH v7 2/12 i3c: master: Use unified device property interface

In reply to Akhil R

Replace all OF-specific functions with unified device property functions as a prerequisite to support both ACPI and device tree.

---
 drivers/i3c/master.c       | 77 +++++++++++++++++++++-----------------
 include/linux/i3c/master.h |  5 ++-
 2 files changed, 46 insertions(+), 36 deletions(-)
drivers/i3c/master.c +43 -34
index f1be38a640ca..4b3d9628bc39 100644
@@ -13,10 +13,12 @@
 #include <linux/dma-mapping.h>
 #include <linux/err.h>
 #include <linux/export.h>
+#include <linux/i2c.h>
 #include <linux/kernel.h>
 #include <linux/list.h>
 #include <linux/of.h>
 #include <linux/pm_runtime.h>
+#include <linux/property.h>
 #include <linux/slab.h>
 #include <linux/spinlock.h>
 #include <linux/workqueue.h>
@@ -491,7 +493,7 @@ static void i3c_bus_cleanup(struct i3c_bus *i3cbus)
 	mutex_unlock(&i3c_core_lock);
 }
 
-static int i3c_bus_init(struct i3c_bus *i3cbus, struct device_node *np)
+static int i3c_bus_init(struct i3c_bus *i3cbus, struct fwnode_handle *fwnode)
 {
 	int ret, start, end, id = -1;
 
@@ -501,8 +503,8 @@ static int i3c_bus_init(struct i3c_bus *i3cbus, struct device_node *np)
 	i3c_bus_init_addrslots(i3cbus);
 	i3cbus->mode = I3C_BUS_MODE_PURE;
 
-	if (np)
-		id = of_alias_get_id(np, "i3c");
+	if (fwnode && is_of_node(fwnode))
+		id = of_alias_get_id(to_of_node(fwnode), "i3c");
 
 	mutex_lock(&i3c_core_lock);
 	if (id >= 0) {
@@ -837,7 +839,7 @@ static void i3c_masterdev_release(struct device *dev)
 	WARN_ON(!list_empty(&bus->devs.i2c) || !list_empty(&bus->devs.i3c));
 	i3c_bus_cleanup(bus);
 
-	of_node_put(dev->of_node);
+	fwnode_handle_put(dev->fwnode);
 }
 
 static const struct device_type i3c_masterdev_type = {
@@ -1044,7 +1046,7 @@ static void i3c_device_release(struct device *dev)
 
 	WARN_ON(i3cdev->desc);
 
-	of_node_put(i3cdev->dev.of_node);
+	fwnode_handle_put(dev->fwnode);
 	kfree(i3cdev);
 }
 
@@ -1928,7 +1930,7 @@ i3c_master_register_new_i3c_devs(struct i3c_master_controller *master)
 			     desc->info.pid);
 
 		if (desc->boardinfo)
-			desc->dev->dev.of_node = desc->boardinfo->of_node;
+			device_set_node(&desc->dev->dev, desc->boardinfo->fwnode);
 
 		ret = device_register(&desc->dev->dev);
 		if (ret) {
@@ -2620,8 +2622,8 @@ EXPORT_SYMBOL_GPL(i3c_master_do_daa);
 #define OF_I3C_REG1_IS_I2C_DEV			BIT(31)
 
 static int
-of_i3c_master_add_i2c_boardinfo(struct i3c_master_controller *master,
-				struct device_node *node, u32 *reg)
+i3c_master_add_i2c_boardinfo(struct i3c_master_controller *master,
+			     struct fwnode_handle *fwnode, u32 *reg)
 {
 	struct i2c_dev_boardinfo *boardinfo;
 	struct device *dev = &master->dev;
@@ -2631,9 +2633,13 @@ of_i3c_master_add_i2c_boardinfo(struct i3c_master_controller *master,
 	if (!boardinfo)
 		return -ENOMEM;
 
-	ret = of_i2c_get_board_info(dev, node, &boardinfo->base);
-	if (ret)
-		return ret;
+	if (is_of_node(fwnode)) {
+		ret = of_i2c_get_board_info(dev, to_of_node(fwnode), &boardinfo->base);
+		if (ret)
+			return ret;
+	} else {
+		return -EINVAL;
+	}
 
 	/*
 	 * The I3C Specification does not clearly say I2C devices with 10-bit
@@ -2649,14 +2655,14 @@ of_i3c_master_add_i2c_boardinfo(struct i3c_master_controller *master,
 	boardinfo->lvr = reg[2];
 
 	list_add_tail(&boardinfo->node, &master->boardinfo.i2c);
-	of_node_get(node);
+	fwnode_handle_get(fwnode);
 
 	return 0;
 }
 
 static int
-of_i3c_master_add_i3c_boardinfo(struct i3c_master_controller *master,
-				struct device_node *node, u32 *reg)
+i3c_master_add_i3c_boardinfo(struct i3c_master_controller *master,
+			     struct fwnode_handle *fwnode, u32 *reg)
 {
 	struct i3c_dev_boardinfo *boardinfo;
 	struct device *dev = &master->dev;
@@ -2679,7 +2685,7 @@ of_i3c_master_add_i3c_boardinfo(struct i3c_master_controller *master,
 
 	boardinfo->static_addr = reg[0];
 
-	if (!of_property_read_u32(node, "assigned-address", &init_dyn_addr)) {
+	if (!fwnode_property_read_u32(fwnode, "assigned-address", &init_dyn_addr)) {
 		if (init_dyn_addr > I3C_MAX_ADDR)
 			return -EINVAL;
 
@@ -2696,14 +2702,14 @@ of_i3c_master_add_i3c_boardinfo(struct i3c_master_controller *master,
 		return -EINVAL;
 
 	boardinfo->init_dyn_addr = init_dyn_addr;
-	boardinfo->of_node = of_node_get(node);
+	boardinfo->fwnode = fwnode_handle_get(fwnode);
 	list_add_tail(&boardinfo->node, &master->boardinfo.i3c);
 
 	return 0;
 }
 
-static int of_i3c_master_add_dev(struct i3c_master_controller *master,
-				 struct device_node *node)
+static int i3c_master_add_dev(struct i3c_master_controller *master,
+			      struct fwnode_handle *fwnode)
 {
 	u32 reg[3];
 	int ret;
@@ -2711,7 +2717,7 @@ static int of_i3c_master_add_dev(struct i3c_master_controller *master,
 	if (!master)
 		return -EINVAL;
 
-	ret = of_property_read_u32_array(node, "reg", reg, ARRAY_SIZE(reg));
+	ret = fwnode_property_read_u32_array(fwnode, "reg", reg, ARRAY_SIZE(reg));
 	if (ret)
 		return ret;
 
@@ -2720,25 +2726,25 @@ static int of_i3c_master_add_dev(struct i3c_master_controller *master,
 	 * dealing with an I2C device.
 	 */
 	if (!reg[1])
-		ret = of_i3c_master_add_i2c_boardinfo(master, node, reg);
+		ret = i3c_master_add_i2c_boardinfo(master, fwnode, reg);
 	else
-		ret = of_i3c_master_add_i3c_boardinfo(master, node, reg);
+		ret = i3c_master_add_i3c_boardinfo(master, fwnode, reg);
 
 	return ret;
 }
 
-static int of_populate_i3c_bus(struct i3c_master_controller *master)
+static int fwnode_populate_i3c_bus(struct i3c_master_controller *master)
 {
 	struct device *dev = &master->dev;
-	struct device_node *i3cbus_np = dev->of_node;
+	struct fwnode_handle *fwnode = dev_fwnode(dev);
 	int ret;
 	u32 val;
 
-	if (!i3cbus_np)
+	if (!fwnode)
 		return 0;
 
-	for_each_available_child_of_node_scoped(i3cbus_np, node) {
-		ret = of_i3c_master_add_dev(master, node);
+	fwnode_for_each_available_child_node_scoped(fwnode, child) {
+		ret = i3c_master_add_dev(master, child);
 		if (ret)
 			return ret;
 	}
@@ -2748,10 +2754,10 @@ static int of_populate_i3c_bus(struct i3c_master_controller *master)
 	 * on the bus are not supporting typical rates, or if the bus topology
 	 * prevents it from using max possible rate.
 	 */
-	if (!of_property_read_u32(i3cbus_np, "i2c-scl-hz", &val))
+	if (!device_property_read_u32(dev, "i2c-scl-hz", &val))
 		master->bus.scl_rate.i2c = val;
 
-	if (!of_property_read_u32(i3cbus_np, "i3c-scl-hz", &val))
+	if (!device_property_read_u32(dev, "i3c-scl-hz", &val))
 		master->bus.scl_rate.i3c = val;
 
 	return 0;
@@ -2806,7 +2812,7 @@ static u8 i3c_master_i2c_get_lvr(struct i2c_client *client)
 	u8 lvr = I3C_LVR_I2C_INDEX(2) | I3C_LVR_I2C_FM_MODE;
 	u32 reg[3];
 
-	if (!of_property_read_u32_array(client->dev.of_node, "reg", reg, ARRAY_SIZE(reg)))
+	if (!fwnode_property_read_u32_array(client->dev.fwnode, "reg", reg, ARRAY_SIZE(reg)))
 		lvr = reg[2];
 
 	return lvr;
@@ -2925,7 +2931,8 @@ static int i3c_master_i2c_adapter_init(struct i3c_master_controller *master)
 	struct i2c_adapter *adap = i3c_master_to_i2c_adapter(master);
 	struct i2c_dev_desc *i2cdev;
 	struct i2c_dev_boardinfo *i2cboardinfo;
-	int ret, id;
+	struct fwnode_handle *fwnode = dev_fwnode(&master->dev);
+	int ret, id = -1;
 
 	adap->dev.parent = master->dev.parent;
 	adap->owner = master->dev.parent->driver->owner;
@@ -2934,7 +2941,9 @@ static int i3c_master_i2c_adapter_init(struct i3c_master_controller *master)
 	adap->timeout = HZ;
 	adap->retries = 3;
 
-	id = of_alias_get_id(master->dev.of_node, "i2c");
+	if (fwnode && is_of_node(fwnode))
+		id = of_alias_get_id(to_of_node(fwnode), "i2c");
+
 	if (id >= 0) {
 		adap->nr = id;
 		ret = i2c_add_numbered_adapter(adap);
@@ -3235,7 +3244,7 @@ int i3c_master_register(struct i3c_master_controller *master,
 		return ret;
 
 	master->dev.parent = parent;
-	master->dev.of_node = of_node_get(parent->of_node);
+	device_set_node(&master->dev, fwnode_handle_get(dev_fwnode(parent)));
 	master->dev.bus = &i3c_bus_type;
 	master->dev.type = &i3c_masterdev_type;
 	master->dev.release = i3c_masterdev_release;
@@ -3254,13 +3263,13 @@ int i3c_master_register(struct i3c_master_controller *master,
 	master->dev.coherent_dma_mask = parent->coherent_dma_mask;
 	master->dev.dma_parms = parent->dma_parms;
 
-	ret = i3c_bus_init(i3cbus, master->dev.of_node);
+	ret = i3c_bus_init(i3cbus, dev_fwnode(&master->dev));
 	if (ret)
 		goto err_put_dev;
 
 	dev_set_name(&master->dev, "i3c-%d", i3cbus->id);
 
-	ret = of_populate_i3c_bus(master);
+	ret = fwnode_populate_i3c_bus(master);
 	if (ret)
 		goto err_put_dev;
 
include/linux/i3c/master.h +3 -2
index 4d2a68793324..a16deb04b2e1 100644
@@ -177,7 +177,8 @@ struct i3c_device_ibi_info {
  * @pid: I3C Provisioned ID exposed by the device. This is a unique identifier
  *	 that may be used to attach boardinfo to i3c_dev_desc when the device
  *	 does not have a static address
- * @of_node: optional DT node in case the device has been described in the DT
+ * @fwnode: Firmware node (DT or ACPI) in case the device has been
+ *	    described in firmware
  *
  * This structure is used to attach board-level information to an I3C device.
  * Not all I3C devices connected on the bus will have a boardinfo. It's only
@@ -189,7 +190,7 @@ struct i3c_dev_boardinfo {
 	u8 init_dyn_addr;
 	u8 static_addr;
 	u64 pid;
-	struct device_node *of_node;
+	struct fwnode_handle *fwnode;
 };
 
 /**
Signature · 1 line
2.43.0
#4 lore raw

Re: [PATCH v7 02/12] i3c: master: Use unified device property interface

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:

New issues:
- [High] Probing of I3C master controllers fails on ACPI platforms if they have child nodes.

Pre-existing issues:
- [High] Dangling pointer left in desc->dev upon device_register() failure leads to Use-After-Free and Double Free on teardown. --

--- Patch [2]: [PATCH v7 02/12] i3c: master: Use unified device property interface --- commit 67a3b30ddd40ca8910b28ac5cd83aad57174f7e8
Author: Akhil R <[email protected]>

i3c: master: Use unified device property interface

Replace all OF-specific functions with unified device property functions as a prerequisite to support both ACPI and device tree.

Quoted 4 lines
drivers/i3c/master.c +0 -0
index f1be38a640ca1..4b3d9628bc39b 100644

[ ... ]

drivers/i3c/master.c i3c_master_register_new_i3c_devs(struct i3c_master_controller *master) · line 1936

Quoted 9 lines
drivers/i3c/master.c +1 -1
@@ -1928,7 +1930,7 @@ i3c_master_register_new_i3c_devs(struct i3c_master_controller *master)
 			     desc->info.pid);
 
 		if (desc->boardinfo)
-			desc->dev->dev.of_node = desc->boardinfo->of_node;
+			device_set_node(&desc->dev->dev, desc->boardinfo->fwnode);
 
 		ret = device_register(&desc->dev->dev);
 		if (ret) {

[Severity: High]
This is a pre-existing issue, but it appears desc->dev is left pointing to a freed device if device_register() fails. When put_device() frees the underlying struct i3c_device (via i3c_device_release()), desc->dev isn't set to NULL.

Later, if i3c_master_unregister_i3c_devs() runs during teardown, could this lead to a use-after-free and double-free?

i3c_master_unregister_i3c_devs() {
    ...
    i3c_bus_for_each_i3cdev(&master->bus, i3cdev) {
        if (!i3cdev->dev)
            continue;

        i3cdev->dev->desc = NULL;
        if (device_is_registered(&i3cdev->dev->dev))
            ...
        else
            put_device(&i3cdev->dev->dev);
    }
}

[ ... ]

Quoted 5 lines
drivers/i3c/master.c +2 -2
-static int of_i3c_master_add_dev(struct i3c_master_controller *master,
-				 struct device_node *node)
+static int i3c_master_add_dev(struct i3c_master_controller *master,
+			      struct fwnode_handle *fwnode)
 {

drivers/i3c/master.c

Quoted 10 lines
drivers/i3c/master.c +1 -1
 	u32 reg[3];
 	int ret;
 
 	if (!master)
 		return -EINVAL;
 
-	ret = of_property_read_u32_array(node, "reg", reg, ARRAY_SIZE(reg));
+	ret = fwnode_property_read_u32_array(fwnode, "reg", reg, ARRAY_SIZE(reg));
 	if (ret)
 		return ret;

[Severity: High]
Since this function is now called for ACPI child nodes via fwnode_populate_i3c_bus(), will this cause I3C master probe to fail on ACPI systems?

ACPI nodes typically use _ADR or _CRS instead of a "reg" property. If "reg" is missing, fwnode_property_read_u32_array() returns an error, which propagates up and causes the entire master registration to abort.

Before this patch, the loop in of_populate_i3c_bus() was skipped completely on ACPI systems because dev->of_node was NULL.

Quoted 6 lines
drivers/i3c/master.c +0 -0
 	/*
 	 * The manufacturer ID can't be 0. If that's the case, we're
 	 * dealing with an I2C device.
 	 */
 	if (!reg[1])
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=2
author
#5 lore raw

PATCH v7 3/12 i3c: master: Support ACPI enumeration of child devices

In reply to Akhil R

Although the existing subsystem allows host controllers to register through the ACPI table, it was not possible to describe I3C or I2C devices when using ACPI. This is because the driver relied on the reg property to retrieve the PID, static address, etc., whereas ACPI uses _ADR or serial resources to describe such devices.

Read _ADR and LVR from ACPI resources and extract the data as per the ACPI specification for an I3C bus. Also read mipi-i3c-static-address as per the MIPI DISCO specifications [1] to get the static address to be used.

Enable describing I3C or I2C devices in the ACPI table. This is required if the device uses a static address or if it needs device-specific properties.

[1] https://www.mipi.org/mipi-disco-for-i3c-download

---
 drivers/i3c/master.c | 151 ++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 143 insertions(+), 8 deletions(-)
drivers/i3c/master.c +143 -8
index 4b3d9628bc39..623c6b2247d9 100644
@@ -5,6 +5,7 @@
  * Author: Boris Brezillon <[email protected]>
  */
 
+#include <linux/acpi.h>
 #include <linux/atomic.h>
 #include <linux/bitmap.h>
 #include <linux/bug.h>
@@ -2621,6 +2622,55 @@ EXPORT_SYMBOL_GPL(i3c_master_do_daa);
 
 #define OF_I3C_REG1_IS_I2C_DEV			BIT(31)
 
+#ifdef CONFIG_ACPI
+static int i3c_acpi_get_i2c_resource(struct acpi_resource *ares, void *data)
+{
+	struct i2c_dev_boardinfo *boardinfo = data;
+	struct acpi_resource_i2c_serialbus *sb;
+
+	if (boardinfo->base.addr || !i2c_acpi_get_i2c_resource(ares, &sb))
+		return 1;
+
+	boardinfo->base.addr = sb->slave_address;
+	if (sb->access_mode == ACPI_I2C_10BIT_MODE)
+		boardinfo->base.flags |= I2C_CLIENT_TEN;
+
+	boardinfo->lvr = sb->lvr;
+
+	return 1;
+}
+
+static int i3c_acpi_add_i2c_boardinfo(struct i2c_dev_boardinfo *boardinfo,
+				      struct fwnode_handle *fwnode)
+{
+	struct acpi_device *adev = to_acpi_device_node(fwnode);
+	LIST_HEAD(resources);
+	int ret;
+
+	boardinfo->base.fwnode = acpi_fwnode_handle(adev);
+	acpi_set_modalias(adev, dev_name(&adev->dev), boardinfo->base.type,
+			  sizeof(boardinfo->base.type));
+
+	ret = acpi_dev_get_resources(adev, &resources,
+				     i3c_acpi_get_i2c_resource, boardinfo);
+	if (ret < 0)
+		return ret;
+
+	acpi_dev_free_resource_list(&resources);
+
+	if (!boardinfo->base.addr)
+		return -ENODEV;
+
+	return 0;
+}
+#else
+static inline int i3c_acpi_add_i2c_boardinfo(struct i2c_dev_boardinfo *boardinfo,
+					     struct fwnode_handle *fwnode)
+{
+	return -ENODEV;
+}
+#endif
+
 static int
 i3c_master_add_i2c_boardinfo(struct i3c_master_controller *master,
 			     struct fwnode_handle *fwnode, u32 *reg)
@@ -2637,6 +2687,15 @@ i3c_master_add_i2c_boardinfo(struct i3c_master_controller *master,
 		ret = of_i2c_get_board_info(dev, to_of_node(fwnode), &boardinfo->base);
 		if (ret)
 			return ret;
+
+		/* LVR is encoded in reg[2] for Device Tree. */
+		boardinfo->lvr = reg[2];
+	} else if (is_acpi_device_node(fwnode)) {
+		ret = i3c_acpi_add_i2c_boardinfo(boardinfo, fwnode);
+		if (ret) {
+			devm_kfree(dev, boardinfo);
+			return ret;
+		}
 	} else {
 		return -EINVAL;
 	}
@@ -2651,9 +2710,6 @@ i3c_master_add_i2c_boardinfo(struct i3c_master_controller *master,
 		return -EOPNOTSUPP;
 	}
 
-	/* LVR is encoded in reg[2]. */
-	boardinfo->lvr = reg[2];
-
 	list_add_tail(&boardinfo->node, &master->boardinfo.i2c);
 	fwnode_handle_get(fwnode);
 
@@ -2708,8 +2764,8 @@ i3c_master_add_i3c_boardinfo(struct i3c_master_controller *master,
 	return 0;
 }
 
-static int i3c_master_add_dev(struct i3c_master_controller *master,
-			      struct fwnode_handle *fwnode)
+static int i3c_master_add_of_dev(struct i3c_master_controller *master,
+				 struct fwnode_handle *fwnode)
 {
 	u32 reg[3];
 	int ret;
@@ -2733,6 +2789,74 @@ static int i3c_master_add_dev(struct i3c_master_controller *master,
 	return ret;
 }
 
+#ifdef CONFIG_ACPI
+static int i3c_master_add_acpi_dev(struct i3c_master_controller *master,
+				   struct fwnode_handle *fwnode)
+{
+	struct acpi_device *adev = to_acpi_device_node(fwnode);
+	acpi_bus_address adr;
+	u32 reg[3] = { 0 };
+	int ret;
+
+	/*
+	 * If the ACPI table entry has _ADR method, it's an I3C device.
+	 * Otherwise it may be an I2C device described by an I2cSerialBus
+	 * resource. If no I2cSerialBus resource is found, ignore the entry.
+	 */
+	if (!acpi_has_method(adev->handle, "_ADR")) {
+		ret = i3c_master_add_i2c_boardinfo(master, fwnode, reg);
+		if (ret == -ENODEV)
+			return 0;
+
+		return ret;
+	}
+
+	adr = acpi_device_adr(adev);
+
+	/* For I3C devices, _ADR will have the 48 bit PID of the device  */
+	reg[1] = upper_32_bits(adr);
+	reg[2] = lower_32_bits(adr);
+
+	fwnode_property_read_u32(fwnode, "mipi-i3c-static-address", &reg[0]);
+
+	return i3c_master_add_i3c_boardinfo(master, fwnode, reg);
+}
+
+static u8 i3c_acpi_i2c_get_lvr(struct i2c_client *client)
+{
+	struct acpi_device *adev = to_acpi_device_node(client->dev.fwnode);
+	struct i2c_dev_boardinfo boardinfo = {};
+	LIST_HEAD(resources);
+	int ret;
+	u8 lvr;
+
+	lvr = I3C_LVR_I2C_INDEX(2) | I3C_LVR_I2C_FM_MODE;
+
+	ret = acpi_dev_get_resources(adev, &resources,
+				     i3c_acpi_get_i2c_resource, &boardinfo);
+	if (ret < 0)
+		return lvr;
+
+	if (boardinfo.base.addr)
+		lvr = boardinfo.lvr;
+
+	acpi_dev_free_resource_list(&resources);
+
+	return lvr;
+}
+#else
+static inline int i3c_master_add_acpi_dev(struct i3c_master_controller *master,
+					  struct fwnode_handle *fwnode)
+{
+	return -ENODEV;
+}
+
+static inline u8 i3c_acpi_i2c_get_lvr(struct i2c_client *client)
+{
+	return I3C_LVR_I2C_INDEX(2) | I3C_LVR_I2C_FM_MODE;
+}
+#endif
+
 static int fwnode_populate_i3c_bus(struct i3c_master_controller *master)
 {
 	struct device *dev = &master->dev;
@@ -2744,7 +2868,13 @@ static int fwnode_populate_i3c_bus(struct i3c_master_controller *master)
 		return 0;
 
 	fwnode_for_each_available_child_node_scoped(fwnode, child) {
-		ret = i3c_master_add_dev(master, child);
+		if (is_of_node(child))
+			ret = i3c_master_add_of_dev(master, child);
+		else if (is_acpi_device_node(child))
+			ret = i3c_master_add_acpi_dev(master, child);
+		else
+			continue;
+
 		if (ret)
 			return ret;
 	}
@@ -2812,8 +2942,13 @@ static u8 i3c_master_i2c_get_lvr(struct i2c_client *client)
 	u8 lvr = I3C_LVR_I2C_INDEX(2) | I3C_LVR_I2C_FM_MODE;
 	u32 reg[3];
 
-	if (!fwnode_property_read_u32_array(client->dev.fwnode, "reg", reg, ARRAY_SIZE(reg)))
-		lvr = reg[2];
+	if (is_of_node(client->dev.fwnode)) {
+		if (!fwnode_property_read_u32_array(client->dev.fwnode, "reg",
+						    reg, ARRAY_SIZE(reg)))
+			lvr = reg[2];
+	} else if (is_acpi_device_node(client->dev.fwnode)) {
+		lvr = i3c_acpi_i2c_get_lvr(client);
+	}
 
 	return lvr;
 }
Signature · 1 line
2.43.0
#6 lore raw

Re: [PATCH v7 03/12] i3c: master: Support ACPI enumeration of child devices

Sashiko has reviewed this patch and found no issues. It looks great!
Quoted 11 lines

Although the existing subsystem allows host controllers to register through the ACPI table, it was not possible to describe I3C or I2C devices when using ACPI. This is because the driver relied on the reg property to retrieve the PID, static address, etc., whereas ACPI uses _ADR or serial resources to describe such devices.

Read _ADR and LVR from ACPI resources and extract the data as per the ACPI specification for an I3C bus. Also read mipi-i3c-static-address as per the MIPI DISCO specifications [1] to get the static address to be used.

Quoted 8 lines

Enable describing I3C or I2C devices in the ACPI table. This is required if the device uses a static address or if it needs device-specific properties.

[1] https://www.mipi.org/mipi-disco-for-i3c-download

Sashiko has reviewed this patch and found no issues. It looks great!

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=3
author
#7 lore raw

PATCH v7 4/12 i3c: master: Add support for devices using SETAASA

In reply to Akhil R

Add support for devices using SETAASA, such as SPD5118 and SPD5108 attached to DDR5 memory modules that do not support ENTDAA. Follow the guidelines proposed by the MIPI Discovery and Configuration Specification [1] for discovering such devices.

SETAASA (Set All Addresses to Static Address) differs from standard I3C address assignment that uses ENTDAA or SETDASA to assign dynamic addresses. Devices using SETAASA assign their pre-defined static addresses as their dynamic addresses during DAA, and it is not mandatory for these devices to implement standard CCC commands like GETPID, GETDCR, or GETBCR. For such devices, it is generally recommended to issue SETHID (specified by JEDEC JESD300) as a prerequisite for SETAASA to stop HID bit flipping.

[1] https://www.mipi.org/mipi-disco-for-i3c-download

---
 drivers/i3c/master.c       | 114 ++++++++++++++++++++++++++++++++++++-
 include/linux/i3c/ccc.h    |   1 +
 include/linux/i3c/master.h |  15 +++++
 3 files changed, 127 insertions(+), 3 deletions(-)
drivers/i3c/master.c +111 -3
index 623c6b2247d9..87749a739e62 100644
@@ -5,6 +5,7 @@
  * Author: Boris Brezillon <[email protected]>
  */
 
+#include <dt-bindings/i3c/i3c.h>
 #include <linux/acpi.h>
 #include <linux/atomic.h>
 #include <linux/bitmap.h>
@@ -1102,6 +1103,51 @@ static int i3c_master_rstdaa_locked(struct i3c_master_controller *master,
 	return ret;
 }
 
+/**
+ * i3c_master_setaasa_locked() - start a SETAASA procedure (Set All Addresses to Static Address)
+ * @master: I3C master object
+ *
+ * Send a SETAASA CCC command to set all attached I3C devices' dynamic addresses to
+ * their static address.
+ *
+ * This function must be called with the bus lock held in write mode.
+ *
+ * First, the SETHID CCC command is sent, followed by the SETAASA CCC.
+ *
+ * Return: 0 in case of success, a positive I3C error code if the error is
+ * one of the official Mx error codes, and a negative error code otherwise.
+ */
+static int i3c_master_setaasa_locked(struct i3c_master_controller *master)
+{
+	struct i3c_ccc_cmd_dest dest;
+	struct i3c_ccc_cmd cmd;
+	int ret;
+
+	/*
+	 * Send SETHID CCC command. Though it is a standard CCC command specified
+	 * in JESD300-5, we are not defining a separate macro to be explicit that
+	 * the value falls under the vendor specific range.
+	 */
+	i3c_ccc_cmd_dest_init(&dest, I3C_BROADCAST_ADDR, 0);
+	i3c_ccc_cmd_init(&cmd, false, I3C_CCC_VENDOR(0, true), &dest, 1);
+	ret = i3c_master_send_ccc_cmd_locked(master, &cmd);
+	i3c_ccc_cmd_dest_cleanup(&dest);
+	if (ret && cmd.err == I3C_ERROR_M2)
+		ret = 0;
+	if (ret)
+		return ret;
+
+	/* Send SETAASA CCC command */
+	i3c_ccc_cmd_dest_init(&dest, I3C_BROADCAST_ADDR, 0);
+	i3c_ccc_cmd_init(&cmd, false, I3C_CCC_SETAASA, &dest, 1);
+	ret = i3c_master_send_ccc_cmd_locked(master, &cmd);
+	i3c_ccc_cmd_dest_cleanup(&dest);
+	if (ret && cmd.err == I3C_ERROR_M2)
+		ret = 0;
+
+	return ret;
+}
+
 /**
  * i3c_master_entdaa_locked() - start a DAA (Dynamic Address Assignment)
  *				procedure
@@ -1869,8 +1915,10 @@ static int i3c_master_early_i3c_dev_add(struct i3c_master_controller *master,
 	int ret;
 
 	i3cdev = i3c_master_alloc_i3c_dev(master, &info);
-	if (IS_ERR(i3cdev))
-		return -ENOMEM;
+	if (IS_ERR(i3cdev)) {
+		ret = -ENOMEM;
+		goto err_reserve_addr;
+	}
 
 	i3cdev->boardinfo = boardinfo;
 
@@ -1878,6 +1926,22 @@ static int i3c_master_early_i3c_dev_add(struct i3c_master_controller *master,
 	if (ret)
 		goto err_free_dev;
 
+	/*
+	 * For devices using SETAASA instead of ENTDAA, the address is statically
+	 * assigned. Update the dynamic address to the provided static address.
+	 * Reattach the I3C device after updating the dynamic address with the same
+	 * static address. It is not mandatory for such devices to implement CCC
+	 * commands like GETPID, GETDCR etc. Hence, we can return after reattaching.
+	 */
+	if (i3cdev->boardinfo->static_addr_method & I3C_ADDR_METHOD_SETAASA) {
+		i3cdev->info.dyn_addr = i3cdev->boardinfo->static_addr;
+		ret = i3c_master_reattach_i3c_dev_locked(i3cdev, 0);
+		if (ret)
+			goto err_detach_dev;
+
+		return 0;
+	}
+
 	ret = i3c_master_setdasa_locked(master, i3cdev->info.static_addr,
 					i3cdev->boardinfo->init_dyn_addr);
 	if (ret)
@@ -1900,6 +1964,16 @@ static int i3c_master_early_i3c_dev_add(struct i3c_master_controller *master,
 	i3c_master_detach_i3c_dev(i3cdev);
 err_free_dev:
 	i3c_master_free_i3c_dev(i3cdev);
+err_reserve_addr:
+	/*
+	 * A target using SETAASA may still get the static address on the
+	 * SETAASA broadcast even if attach fails here. Keep the address
+	 * reserved so that it is not assigned to another device during DAA.
+	 */
+	if (boardinfo->static_addr_method & I3C_ADDR_METHOD_SETAASA)
+		i3c_bus_set_addr_slot_status(&master->bus,
+					     boardinfo->static_addr,
+					     I3C_ADDR_SLOT_RSVD);
 
 	return ret;
 }
@@ -2272,6 +2346,19 @@ static int i3c_master_bus_init(struct i3c_master_controller *master)
 			i3c_master_early_i3c_dev_add(master, i3cboardinfo);
 	}
 
+	/*
+	 * SETAASA is a broadcast CCC. Issue it after SETDASA so that devices
+	 * configured for SETDASA (or supporting both methods) are assigned
+	 * first, matching MIPI DISCO guidance to prefer SETDASA when both are
+	 * available. Targets that already have a dynamic address ignore the
+	 * later SETAASA broadcast.
+	 */
+	if (master->addr_method & I3C_ADDR_METHOD_SETAASA) {
+		ret = i3c_master_setaasa_locked(master);
+		if (ret)
+			goto err_rstdaa;
+	}
+
 	ret = i3c_master_do_daa(master);
 	if (ret)
 		goto err_rstdaa;
@@ -2723,7 +2810,7 @@ i3c_master_add_i3c_boardinfo(struct i3c_master_controller *master,
 	struct i3c_dev_boardinfo *boardinfo;
 	struct device *dev = &master->dev;
 	enum i3c_addr_slot_status addrstatus;
-	u32 init_dyn_addr = 0;
+	u32 init_dyn_addr = 0, static_addr_method = 0;
 
 	boardinfo = devm_kzalloc(dev, sizeof(*boardinfo), GFP_KERNEL);
 	if (!boardinfo)
@@ -2741,7 +2828,19 @@ i3c_master_add_i3c_boardinfo(struct i3c_master_controller *master,
 
 	boardinfo->static_addr = reg[0];
 
+	if (!fwnode_property_read_u32(fwnode, "mipi-i3c-static-method", &static_addr_method))
+		boardinfo->static_addr_method = static_addr_method &
+					(I3C_ADDR_METHOD_SETDASA | I3C_ADDR_METHOD_SETAASA);
+
 	if (!fwnode_property_read_u32(fwnode, "assigned-address", &init_dyn_addr)) {
+		/*
+		 * When a device advertises both SETDASA and SETAASA, an explicit
+		 * dynamic address selects SETDASA (MIPI DISCO prefers it); drop
+		 * SETAASA so it is not used for this device.
+		 */
+		if (boardinfo->static_addr_method & I3C_ADDR_METHOD_SETDASA)
+			boardinfo->static_addr_method &= ~I3C_ADDR_METHOD_SETAASA;
+
 		if (init_dyn_addr > I3C_MAX_ADDR)
 			return -EINVAL;
 
@@ -2751,6 +2850,14 @@ i3c_master_add_i3c_boardinfo(struct i3c_master_controller *master,
 			return -EINVAL;
 	}
 
+	if (boardinfo->static_addr_method & I3C_ADDR_METHOD_SETAASA) {
+		/* For SETAASA, static address is taken as the dynamic address. */
+		init_dyn_addr = boardinfo->static_addr;
+	}
+
+	/* Update the address methods required for device discovery */
+	master->addr_method |= boardinfo->static_addr_method;
+
 	boardinfo->pid = ((u64)reg[1] << 32) | reg[2];
 
 	if ((boardinfo->pid & GENMASK_ULL(63, 48)) ||
@@ -3385,6 +3492,7 @@ int i3c_master_register(struct i3c_master_controller *master,
 	master->dev.release = i3c_masterdev_release;
 	master->ops = ops;
 	master->secondary = secondary;
+	master->addr_method = I3C_ADDR_METHOD_SETDASA;
 	INIT_LIST_HEAD(&master->boardinfo.i2c);
 	INIT_LIST_HEAD(&master->boardinfo.i3c);
 
include/linux/i3c/ccc.h +1 -0
index ad59a4ae60d1..a145d766ab6f 100644
@@ -32,6 +32,7 @@
 #define I3C_CCC_DEFSLVS			I3C_CCC_ID(0x8, true)
 #define I3C_CCC_ENTTM			I3C_CCC_ID(0xb, true)
 #define I3C_CCC_ENTHDR(x)		I3C_CCC_ID(0x20 + (x), true)
+#define I3C_CCC_SETAASA			I3C_CCC_ID(0x29, true)
 
 /* Unicast-only commands */
 #define I3C_CCC_SETDASA			I3C_CCC_ID(0x7, false)
include/linux/i3c/master.h +15 -0
index a16deb04b2e1..2dc139a217bf 100644
@@ -174,6 +174,14 @@ struct i3c_device_ibi_info {
  *		 assigned a dynamic address by the master. Will be used during
  *		 bus initialization to assign it a specific dynamic address
  *		 before starting DAA (Dynamic Address Assignment)
+ * @static_addr_method: Bitmap describing which methods of Dynamic Address
+ *		 Assignment from a Static Address are supported by this I3C Target.
+ *		 A value of 1 in a bit position indicates that the I3C target
+ *		 supports that method, and a value of 0 indicates that the I3C
+ *		 target does not support that method.
+ *		 Bit 0: SETDASA
+ *		 Bit 1: SETAASA
+ *		 All other bits are reserved.
  * @pid: I3C Provisioned ID exposed by the device. This is a unique identifier
  *	 that may be used to attach boardinfo to i3c_dev_desc when the device
  *	 does not have a static address
@@ -189,6 +197,7 @@ struct i3c_dev_boardinfo {
 	struct list_head node;
 	u8 init_dyn_addr;
 	u8 static_addr;
+	u8 static_addr_method;
 	u64 pid;
 	struct fwnode_handle *fwnode;
 };
@@ -517,6 +526,11 @@ struct i3c_master_controller_ops {
  * @boardinfo.i2c: list of I2C boardinfo objects
  * @boardinfo: board-level information attached to devices connected on the bus
  * @bus: I3C bus exposed by this master
+ * @addr_method: Bitmap describing which methods of Address Assignment required
+ *		 to be run for discovering all the devices on the bus.
+ *		 Bit 0: SETDASA
+ *		 Bit 1: SETAASA
+ *		 All other bits are reserved.
  * @wq: freezable workqueue which can be used by master
  *	drivers if they need to postpone operations that need to take place
  *	in a thread context. Typical examples are Hot Join processing which
@@ -552,6 +566,7 @@ struct i3c_master_controller {
 		struct list_head i2c;
 	} boardinfo;
 	struct i3c_bus bus;
+	u8 addr_method;
 	struct workqueue_struct *wq;
 	struct work_struct hj_work;
 	struct work_struct reg_work;
Signature · 1 line
2.43.0
#8 lore raw

Re: [PATCH v7 04/12] i3c: master: Add support for devices using SETAASA

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider: - [High] Bypassing PID validation for SETAASA devices permits identical dummy PIDs, causing sysfs name…

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
- [High] Bypassing PID validation for SETAASA devices permits identical dummy PIDs, causing sysfs name collisions during registration and triggering a Use-After-Free.
- [High] Sending SETAASA and SETHID CCCs will unconditionally fail bus initialization on most hardware master controllers.
- [High] SETAASA devices become permanently unreachable after system hibernation/restore because the SETAASA broadcast is not re-sent.
--

--- Patch [4]: [PATCH v7 04/12] i3c: master: Add support for devices using SETAASA --- commit 61911732853b10649effe5164454ca8a4a87e970
Author: Akhil R <[email protected]>
Subject: i3c: master: Add support for devices using SETAASA

This patch adds support for discovering and configuring devices using the SETAASA method, which statically assigns addresses without using ENTDAA. It adheres to the MIPI Discovery and Configuration Specification.

Quoted 4 lines
drivers/i3c/master.c +0 -0
index 623c6b2247d9f..87749a739e62d 100644

[ ... ]

Quoted 9 lines
drivers/i3c/master.c +5 -0
@@ -1878,6 +1926,22 @@ static int i3c_master_early_i3c_dev_add(struct i3c_master_controller *master,
 	if (ret)
 		goto err_free_dev;
 
+	/*
+	 * For devices using SETAASA instead of ENTDAA, the address is statically
+	 * assigned. Update the dynamic address to the provided static address.
+	 * Reattach the I3C device after updating the dynamic address with the same
+	 * static address. It is not mandatory for such devices to implement CCC

drivers/i3c/master.c static int i3c_master_early_i3c_dev_add(struct i3c_master_controller *master, · line 1943

Quoted 10 lines
drivers/i3c/master.c +10 -0
+	 * commands like GETPID, GETDCR etc. Hence, we can return after reattaching.
+	 */
+	if (i3cdev->boardinfo->static_addr_method & I3C_ADDR_METHOD_SETAASA) {
+		i3cdev->info.dyn_addr = i3cdev->boardinfo->static_addr;
+		ret = i3c_master_reattach_i3c_dev_locked(i3cdev, 0);
+		if (ret)
+			goto err_detach_dev;
+
+		return 0;
+	}

[Severity: High]
Does this early return prevent the device from acquiring a unique PID?

Since PID validation is bypassed for SETAASA devices in i3c_master_add_i3c_boardinfo(), multiple devices can share the same default PID (like 0). Because this skips i3c_master_retrieve_dev_info(), the dummy PID is never replaced by a real one.

When i3c_master_register_new_i3c_devs() attempts to register these devices, the sysfs registration fails due to duplicate names. The error handling calls put_device() but leaves a dangling pointer in desc->dev.

Could this dangling pointer lead to a use-after-free when i3c_master_unregister_i3c_devs() later accesses and frees it again?

[ ... ]

Quoted 6 lines
drivers/i3c/master.c +2 -0
@@ -2272,6 +2346,19 @@ static int i3c_master_bus_init(struct i3c_master_controller *master)
 			i3c_master_early_i3c_dev_add(master, i3cboardinfo);
 	}
 
+	/*
+	 * SETAASA is a broadcast CCC. Issue it after SETDASA so that devices

drivers/i3c/master.c static int i3c_master_bus_init(struct i3c_master_controller *master) · line 2360

Quoted 10 lines
drivers/i3c/master.c +10 -0
+	 * configured for SETDASA (or supporting both methods) are assigned
+	 * first, matching MIPI DISCO guidance to prefer SETDASA when both are
+	 * available. Targets that already have a dynamic address ignore the
+	 * later SETAASA broadcast.
+	 */
+	if (master->addr_method & I3C_ADDR_METHOD_SETAASA) {
+		ret = i3c_master_setaasa_locked(master);
+		if (ret)
+			goto err_rstdaa;
+	}

[Severity: High]
What happens here if the underlying master controller driver rejects the SETAASA and SETHID CCC commands?

Most mainline I3C master drivers, like Cadence and DesignWare, implement a strict allowlist in their supports_ccc_cmd callback. If they reject these new commands, i3c_master_send_ccc_cmd_locked() returns -EOPNOTSUPP.

Will this cause the entire I3C bus initialization to abort and detach all devices, effectively breaking all I3C communications for that controller?

[Severity: High]
How are SETAASA devices handled during system resume?

During resume, i3c_master_do_daa_ext() issues an RSTDAA broadcast to reset dynamic addresses and then calls ENTDAA. Since SETAASA devices do not participate in ENTDAA, and SETAASA is not re-broadcasted there, will these devices lose their dynamic address on the physical bus?

If the kernel still retains the i3c_dev_desc with the dynamic address uncleared, won't all subsequent I/O to these targets fail?

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=4
author
#9 lore raw

Re: [PATCH v7 04/12] i3c: master: Add support for devices using SETAASA

Quoted 29 lines · Sashiko AI review

commit 61911732853b10649effe5164454ca8a4a87e970
Author: Akhil R <[email protected]>

i3c: master: Add support for devices using SETAASA

This patch adds support for discovering and configuring devices using the SETAASA method, which statically assigns addresses without using ENTDAA. It adheres to the MIPI Discovery and Configuration Specification.

[ ... ]

drivers/i3c/master.c static int i3c_master_early_i3c_dev_add(struct i3c_master_controller *master, · line 1933

Quoted 9 lines
drivers/i3c/master.c +8 -0
@@ -1878,6 +1926,22 @@ static int i3c_master_early_i3c_dev_add(struct i3c_master_controller *master,
+	if (i3cdev->boardinfo->static_addr_method & I3C_ADDR_METHOD_SETAASA) {
+		i3cdev->info.dyn_addr = i3cdev->boardinfo->static_addr;
+		ret = i3c_master_reattach_i3c_dev_locked(i3cdev, 0);
+		if (ret)
+			goto err_detach_dev;
+
+		return 0;
+	}

[Severity: High]
Does this early return prevent the device from acquiring a unique PID?

Since PID validation is bypassed for SETAASA devices in i3c_master_add_i3c_boardinfo(), multiple devices can share the same default PID (like 0). Because this skips i3c_master_retrieve_dev_info(), the dummy PID is never replaced by a real one.

Quoted 6 lines

When i3c_master_register_new_i3c_devs() attempts to register these devices, the sysfs registration fails due to duplicate names. The error handling calls put_device() but leaves a dangling pointer in desc->dev.

Could this dangling pointer lead to a use-after-free when i3c_master_unregister_i3c_devs() later accesses and frees it again?

As clarified in v6, devices without a PID are named using their static address in the next patch of this series, and two devices cannot share the same static address. The dangling desc->dev on device_register() failure is pre-existing and will have to be fixed separately.

Quoted 13 lines

[ ... ]

drivers/i3c/master.c static int i3c_master_bus_init(struct i3c_master_controller *master) · line 2350

Quoted 6 lines
drivers/i3c/master.c +5 -0
@@ -2272,6 +2346,19 @@ static int i3c_master_bus_init(struct i3c_master_controller *master)
+	if (master->addr_method & I3C_ADDR_METHOD_SETAASA) {
+		ret = i3c_master_setaasa_locked(master);
+		if (ret)
+			goto err_rstdaa;
+	}

[Severity: High]
What happens here if the underlying master controller driver rejects the SETAASA and SETHID CCC commands?

Quoted 6 lines

Most mainline I3C master drivers, like Cadence and DesignWare, implement a strict allowlist in their supports_ccc_cmd callback. If they reject these new commands, i3c_master_send_ccc_cmd_locked() returns -EOPNOTSUPP.

Will this cause the entire I3C bus initialization to abort and detach all devices, effectively breaking all I3C communications for that controller?

This is clarified in v6. It's an opt-in property and existing drivers are unaffected.

Quoted 10 lines

[Severity: High]
How are SETAASA devices handled during system resume?

During resume, i3c_master_do_daa_ext() issues an RSTDAA broadcast to reset dynamic addresses and then calls ENTDAA. Since SETAASA devices do not participate in ENTDAA, and SETAASA is not re-broadcasted there, will these devices lose their dynamic address on the physical bus?

If the kernel still retains the i3c_dev_desc with the dynamic address uncleared, won't all subsequent I/O to these targets fail?

i3c_master_do_daa_ext() is only reached through the mipi-i3c-hci driver when HCI_QUIRK_RPM_ALLOWED is set and not through dw-i3c-master. So this issue does not have any impact today since HCI_QUIRK_RPM_ALLOWED is set only by intel-lpss-i3c platform.

The same issue also applies to other statically configured devices like SETDASA devices that do not implement ENTDAA. It will be better to fix this separately.

Best Regards,
Akhil

author
#10 lore raw

PATCH v7 5/12 i3c: master: Add support for devices without PID

In reply to Akhil R

Devices using SETAASA for address assignment are not required to have a 48-bit PID according to the I3C specification. Allow such devices to register and use the static address where PID was required.

---
 drivers/i3c/master.c | 52 ++++++++++++++++++++++++++++++++++----------
 1 file changed, 41 insertions(+), 11 deletions(-)
drivers/i3c/master.c +41 -11
index 87749a739e62..488177d95e48 100644
@@ -2001,8 +2001,17 @@ i3c_master_register_new_i3c_devs(struct i3c_master_controller *master)
 		desc->dev->dev.type = &i3c_device_type;
 		desc->dev->dev.bus = &i3c_bus_type;
 		desc->dev->dev.release = i3c_device_release;
-		dev_set_name(&desc->dev->dev, "%d-%llx", master->bus.id,
-			     desc->info.pid);
+
+		/*
+		 * For devices without PID (e.g., SETAASA devices), use
+		 * static address for naming instead.
+		 */
+		if (desc->info.pid)
+			dev_set_name(&desc->dev->dev, "%d-%llx", master->bus.id,
+				     desc->info.pid);
+		else
+			dev_set_name(&desc->dev->dev, "%d-%02x", master->bus.id,
+				     desc->info.static_addr);
 
 		if (desc->boardinfo)
 			device_set_node(&desc->dev->dev, desc->boardinfo->fwnode);
@@ -2401,8 +2410,18 @@ static void i3c_master_attach_boardinfo(struct i3c_dev_desc *i3cdev)
 	struct i3c_dev_boardinfo *i3cboardinfo;
 
 	list_for_each_entry(i3cboardinfo, &master->boardinfo.i3c, node) {
-		if (i3cdev->info.pid != i3cboardinfo->pid)
-			continue;
+		/*
+		 * For devices without PID (e.g., SETAASA devices), match by
+		 * static address. For devices with PID, match by PID.
+		 */
+		if (i3cboardinfo->pid) {
+			if (i3cdev->info.pid != i3cboardinfo->pid)
+				continue;
+		} else {
+			if (!i3cboardinfo->static_addr ||
+			    i3cdev->info.static_addr != i3cboardinfo->static_addr)
+				continue;
+		}
 
 		i3cdev->boardinfo = i3cboardinfo;
 		i3cdev->info.static_addr = i3cboardinfo->static_addr;
@@ -2416,8 +2435,12 @@ i3c_master_search_i3c_dev_duplicate(struct i3c_dev_desc *refdev)
 	struct i3c_master_controller *master = i3c_dev_get_master(refdev);
 	struct i3c_dev_desc *i3cdev;
 
+	if (!refdev->info.pid)
+		return NULL;
+
 	i3c_bus_for_each_i3cdev(&master->bus, i3cdev) {
-		if (i3cdev != refdev && i3cdev->info.pid == refdev->info.pid)
+		if (i3cdev != refdev && i3cdev->info.pid &&
+		    i3cdev->info.pid == refdev->info.pid)
 			return i3cdev;
 	}
 
@@ -2860,9 +2883,16 @@ i3c_master_add_i3c_boardinfo(struct i3c_master_controller *master,
 
 	boardinfo->pid = ((u64)reg[1] << 32) | reg[2];
 
-	if ((boardinfo->pid & GENMASK_ULL(63, 48)) ||
-	    I3C_PID_RND_LOWER_32BITS(boardinfo->pid))
-		return -EINVAL;
+	/* For SETAASA devices, validate the static address instead of PID */
+	if (boardinfo->static_addr_method & I3C_ADDR_METHOD_SETAASA) {
+		if (!boardinfo->static_addr)
+			return -EINVAL;
+	} else {
+		if (!I3C_PID_MANUF_ID(boardinfo->pid) ||
+		    (boardinfo->pid & GENMASK_ULL(63, 48)) ||
+		    I3C_PID_RND_LOWER_32BITS(boardinfo->pid))
+			return -EINVAL;
+	}
 
 	boardinfo->init_dyn_addr = init_dyn_addr;
 	boardinfo->fwnode = fwnode_handle_get(fwnode);
@@ -2885,10 +2915,10 @@ static int i3c_master_add_of_dev(struct i3c_master_controller *master,
 		return ret;
 
 	/*
-	 * The manufacturer ID can't be 0. If reg[1] == 0 that means we're
-	 * dealing with an I2C device.
+	 * I3C device should have either the manufacturer ID specified or the
+	 * address discovery method specified. Else treat it as an I2C device.
 	 */
-	if (!reg[1])
+	if (!reg[1] && !fwnode_property_present(fwnode, "mipi-i3c-static-method"))
 		ret = i3c_master_add_i2c_boardinfo(master, fwnode, reg);
 	else
 		ret = i3c_master_add_i3c_boardinfo(master, fwnode, reg);
Signature · 1 line
2.43.0
#11 lore raw

Re: [PATCH v7 05/12] i3c: master: Add support for devices without PID

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:

New issues:
- [High] Dead code in i3c_master_attach_boardinfo fallback match logic prevents Device Tree node attachment for SETAASA devices discovered via DAA.
- [High] Duplicate I3C device creation for SETAASA devices participating in Dynamic Address Assignment.

Pre-existing issues:
- [High] Use-After-Free and Double-Free on driver teardown due to a dangling pointer left by device_register() failure. --

--- Patch [5]: [PATCH v7 05/12] i3c: master: Add support for devices without PID --- commit 978db33da88aedc2a14abd69fdc128b5e3961e92
Author: Akhil R <[email protected]>

i3c: master: Add support for devices without PID

Devices using SETAASA for address assignment are not required to have a 48-bit PID according to the I3C specification. Allow such devices to register and use the static address where PID was required.

Quoted 14 lines
drivers/i3c/master.c +4 -2
index 87749a739e62d..488177d95e484 100644
@@ -2001,8 +2001,17 @@ i3c_master_register_new_i3c_devs(struct i3c_master_controller *master)
 		desc->dev->dev.type = &i3c_device_type;
 		desc->dev->dev.bus = &i3c_bus_type;
 		desc->dev->dev.release = i3c_device_release;
-		dev_set_name(&desc->dev->dev, "%d-%llx", master->bus.id,
-			     desc->info.pid);
+
+		/*
+		 * For devices without PID (e.g., SETAASA devices), use
+		 * static address for naming instead.

drivers/i3c/master.c i3c_master_register_new_i3c_devs(struct i3c_master_controller *master) · line 2017

Quoted 10 lines
drivers/i3c/master.c +7 -0
+		 */
+		if (desc->info.pid)
+			dev_set_name(&desc->dev->dev, "%d-%llx", master->bus.id,
+				     desc->info.pid);
+		else
+			dev_set_name(&desc->dev->dev, "%d-%02x", master->bus.id,
+				     desc->info.static_addr);
 
 		if (desc->boardinfo)
 			device_set_node(&desc->dev->dev, desc->boardinfo->fwnode);

[Severity: High]
This is a pre-existing issue, but does device_register() failure in this function leave a dangling pointer that causes a use-after-free later?

Just below this hunk in i3c_master_register_new_i3c_devs():

    ret = device_register(&desc->dev->dev);
    if (ret) {
        dev_err(&master->dev,
            "Failed to add I3C device (err = %d)\n", ret);
        put_device(&desc->dev->dev);
    }

When put_device() frees desc->dev->dev, desc->dev is not set to NULL. Later, if the master is unregistered, i3c_master_unregister_i3c_devs() iterates through the list and dereferences this dangling pointer:

    if (!i3cdev->dev)
        continue;
    i3cdev->dev->desc = NULL;
    // ...
    else
        put_device(&i3cdev->dev->dev);

Could desc->dev be set to NULL after put_device() to prevent this?

[ ... ]

Quoted 8 lines
drivers/i3c/master.c +2 -2
@@ -2401,8 +2410,18 @@ static void i3c_master_attach_boardinfo(struct i3c_dev_desc *i3cdev)
 	struct i3c_dev_boardinfo *i3cboardinfo;
 
 	list_for_each_entry(i3cboardinfo, &master->boardinfo.i3c, node) {
-		if (i3cdev->info.pid != i3cboardinfo->pid)
-			continue;
+		/*
+		 * For devices without PID (e.g., SETAASA devices), match by

drivers/i3c/master.c static void i3c_master_attach_boardinfo(struct i3c_dev_desc *i3cdev) · line 2424

Quoted 10 lines
drivers/i3c/master.c +10 -0
+		 * static address. For devices with PID, match by PID.
+		 */
+		if (i3cboardinfo->pid) {
+			if (i3cdev->info.pid != i3cboardinfo->pid)
+				continue;
+		} else {
+			if (!i3cboardinfo->static_addr ||
+			    i3cdev->info.static_addr != i3cboardinfo->static_addr)
+				continue;
+		}

[Severity: High]
Will this fallback matching logic fail for SETAASA devices discovered dynamically?

If a SETAASA device without a PID in the Device Tree loses its dynamic address and asserts a Hot-Join, the master will execute Dynamic Address Assignment (DAA).

During DAA, __i3c_master_add_i3c_dev_locked() allocates a new device descriptor and initializes newdev's static_addr to 0, because the static address cannot be retrieved via CCCs.

When i3c_master_attach_boardinfo(newdev) is then called, i3cdev->info.static_addr will be 0. Would this cause the condition i3cdev->info.static_addr != i3cboardinfo->static_addr to always evaluate to true, skipping the boardinfo assignment and preventing the driver from binding?

Quoted 7 lines
drivers/i3c/master.c +0 -0
 		i3cdev->boardinfo = i3cboardinfo;
 		i3cdev->info.static_addr = i3cboardinfo->static_addr;
@@ -2416,8 +2435,12 @@ i3c_master_search_i3c_dev_duplicate(struct i3c_dev_desc *refdev)
 	struct i3c_master_controller *master = i3c_dev_get_master(refdev);
 	struct i3c_dev_desc *i3cdev;
 

drivers/i3c/master.c i3c_master_search_i3c_dev_duplicate(struct i3c_dev_desc *refdev) · line 2446

Quoted 10 lines
drivers/i3c/master.c +5 -1
+	if (!refdev->info.pid)
+		return NULL;
+
 	i3c_bus_for_each_i3cdev(&master->bus, i3cdev) {
-		if (i3cdev != refdev && i3cdev->info.pid == refdev->info.pid)
+		if (i3cdev != refdev && i3cdev->info.pid &&
+		    i3cdev->info.pid == refdev->info.pid)
 			return i3cdev;
 	}
 

[Severity: High]
Does this logic lead to duplicate device creation for SETAASA devices that participate in Dynamic Address Assignment?

If a SETAASA device is defined in the DT without a PID, it is initially added to the master's bus list with pid = 0. If the device later asserts a Hot-Join, the master executes DAA, discovers the device, and retrieves its true physical PID.

In __i3c_master_add_i3c_dev_locked(), the old device is searched using:
olddev = i3c_master_search_i3c_dev_duplicate(newdev);

Since the existing device has pid = 0, the check here for i3cdev->info.pid and i3cdev->info.pid == refdev->info.pid will fail to match the newly discovered PID to the existing device. Will this cause the master to add the newly discovered device instead of detaching and updating the existing one, resulting in duplicate devices on the bus?

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=5
author
#12 lore raw

Re: [PATCH v7 05/12] i3c: master: Add support for devices without PID

Quoted 52 lines · Sashiko AI review

commit 978db33da88aedc2a14abd69fdc128b5e3961e92
Author: Akhil R <[email protected]>

i3c: master: Add support for devices without PID

Devices using SETAASA for address assignment are not required to have a 48-bit PID according to the I3C specification. Allow such devices to register and use the static address where PID was required.

drivers/i3c/master.c

Quoted 4 lines
drivers/i3c/master.c +0 -0
index 87749a739e62d..488177d95e484 100644

[ ... ]

[Severity: High]
This is a pre-existing issue, but does device_register() failure in this function leave a dangling pointer that causes a use-after-free later?

When put_device() frees desc->dev->dev, desc->dev is not set to NULL. Later, if the master is unregistered, i3c_master_unregister_i3c_devs() iterates through the list and dereferences this dangling pointer.

Could desc->dev be set to NULL after put_device() to prevent this? [ ... ]

drivers/i3c/master.c static void i3c_master_attach_boardinfo(struct i3c_dev_desc *i3cdev) · line 2418

Quoted 12 lines
drivers/i3c/master.c +8 -2
@@ -2401,8 +2410,18 @@ static void i3c_master_attach_boardinfo(struct i3c_dev_desc *i3cdev)
 	list_for_each_entry(i3cboardinfo, &master->boardinfo.i3c, node) {
-		if (i3cdev->info.pid != i3cboardinfo->pid)
-			continue;
+		if (i3cboardinfo->pid) {
+			if (i3cdev->info.pid != i3cboardinfo->pid)
+				continue;
+		} else {
+			if (!i3cboardinfo->static_addr ||
+			    i3cdev->info.static_addr != i3cboardinfo->static_addr)
+				continue;
+		}

[Severity: High]
Will this fallback matching logic fail for SETAASA devices discovered dynamically?

If a SETAASA device without a PID in the Device Tree loses its dynamic address and asserts a Hot-Join, the master will execute Dynamic Address Assignment (DAA).

During DAA, __i3c_master_add_i3c_dev_locked() allocates a new device descriptor and initializes newdev's static_addr to 0, because the static address cannot be retrieved via CCCs.

Quoted 5 lines

When i3c_master_attach_boardinfo(newdev) is then called, i3cdev->info.static_addr will be 0. Would this cause the condition i3cdev->info.static_addr != i3cboardinfo->static_addr to always evaluate to true, skipping the boardinfo assignment and preventing the driver from binding?

A device without a PID cannot be discovered through DAA, so this should not happen. Per the MIPI I3C specification, a Target that supports ENTDAA shall have a 48-bit Provisioned ID (Section 5.1.4.1.1), which it transmits during the procedure, and Hot-Join is only compatible with ENTDAA - "Targets that do not support ENTDAA shall not use Hot-Join" (Section 5.1.5). A device that supports both SETAASA and ENTDAA does have a PID and is expected to describe it in DT/ACPI, in which case the PID match applies.

Quoted 22 lines

drivers/i3c/master.c i3c_master_search_i3c_dev_duplicate(struct i3c_dev_desc *refdev) · line 2442

Quoted 12 lines
drivers/i3c/master.c +5 -1
 		i3cdev->boardinfo = i3cboardinfo;
 		i3cdev->info.static_addr = i3cboardinfo->static_addr;
@@ -2416,8 +2435,12 @@ i3c_master_search_i3c_dev_duplicate(struct i3c_dev_desc *refdev)
+	if (!refdev->info.pid)
+		return NULL;
+
 	i3c_bus_for_each_i3cdev(&master->bus, i3cdev) {
-		if (i3cdev != refdev && i3cdev->info.pid == refdev->info.pid)
+		if (i3cdev != refdev && i3cdev->info.pid &&
+		    i3cdev->info.pid == refdev->info.pid)
 			return i3cdev;
 	}

[Severity: High]
Does this logic lead to duplicate device creation for SETAASA devices that participate in Dynamic Address Assignment?

If a SETAASA device is defined in the DT without a PID, it is initially added to the master's bus list with pid = 0. If the device later asserts a Hot-Join, the master executes DAA, discovers the device, and retrieves its true physical PID.

Quoted 5 lines

Since the existing device has pid = 0, the check here for i3cdev->info.pid and i3cdev->info.pid == refdev->info.pid will fail to match the newly discovered PID to the existing device. Will this cause the master to add the newly discovered device instead of detaching and updating the existing one, resulting in duplicate devices on the bus?

This also needs a device without a PID to take part in ENTDAA, which the specification does not allow, as above. A device that already has a Dynamic Address assigned through SETAASA does not participate in ENTDAA either (Section 5.1.4.2).

The !pid checks are intentional in this patch. Devices without a PID have pid == 0 and would otherwise match each other, and one of them would be freed as a duplicate. Devices that do have a PID are matched as before.

Best Regards,
Akhil

author
#13 lore raw

PATCH v7 6/12 i3c: master: match I3C device through DT and ACPI

In reply to Akhil R

SETAASA-based devices cannot always be identified by PID or DCR; the standard I3C id_table matching may not be applicable. Allow such devices to match through Device Tree or ACPI.

---
 drivers/i3c/master.c | 20 +++++++++++++++++++-
 1 file changed, 19 insertions(+), 1 deletion(-)
drivers/i3c/master.c +19 -1
index 488177d95e48..f94e2410d37b 100644
@@ -19,6 +19,7 @@
 #include <linux/kernel.h>
 #include <linux/list.h>
 #include <linux/of.h>
+#include <linux/of_device.h>
 #include <linux/pm_runtime.h>
 #include <linux/property.h>
 #include <linux/slab.h>
@@ -345,15 +346,32 @@ static int i3c_device_match(struct device *dev, const struct device_driver *drv)
 {
 	struct i3c_device *i3cdev;
 	const struct i3c_driver *i3cdrv;
+	u8 static_addr_method = 0;
 
 	if (dev->type != &i3c_device_type)
 		return 0;
 
 	i3cdev = dev_to_i3cdev(dev);
 	i3cdrv = drv_to_i3cdrv(drv);
-	if (i3c_device_match_id(i3cdev, i3cdrv->id_table))
+
+	if (i3cdev->desc && i3cdev->desc->boardinfo)
+		static_addr_method = i3cdev->desc->boardinfo->static_addr_method;
+
+	/*
+	 * SETAASA-based devices need not always have a matching ID since
+	 * it is not mandatory for such devices to implement deviceinfo
+	 * CCC commands. Allow them to register through DT or ACPI.
+	 */
+	if (i3cdrv->id_table && i3c_device_match_id(i3cdev, i3cdrv->id_table))
 		return 1;
 
+	if (static_addr_method & I3C_ADDR_METHOD_SETAASA) {
+		if (of_driver_match_device(dev, drv))
+			return 1;
+		if (acpi_driver_match_device(dev, drv))
+			return 1;
+	}
+
 	return 0;
 }
 
Signature · 1 line
2.43.0
#14 lore raw

Re: [PATCH v7 06/12] i3c: master: match I3C device through DT and ACPI

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:

Pre-existing issues:
- [High] Uninitialized stack memory read and kernel stack information leak during device unregistration when i3cdev->desc is NULL.
- [High] NULL pointer dereference in sysfs attribute show functions when accessed concurrently with device removal. --

--- Patch [6]: [PATCH v7 06/12] i3c: master: match I3C device through DT and ACPI --- commit d545f133fafb1ddd78e75d979b76dc143ac8e0b0
Author: Akhil R <[email protected]>

i3c: master: match I3C device through DT and ACPI

This commit updates i3c_device_match() to allow SETAASA-based I3C devices to be matched through Device Tree or ACPI. It checks if the device has a SETAASA static address method and uses of_driver_match_device() or acpi_driver_match_device() as fallbacks when standard ID matching is not applicable.

Quoted 4 lines
drivers/i3c/master.c +0 -0
index 488177d95e484..f94e2410d37ba 100644

[ ... ]

Quoted 12 lines
drivers/i3c/master.c +1 -1
@@ -345,15 +346,32 @@ static int i3c_device_match(struct device *dev, const struct device_driver *drv)
 {
 	struct i3c_device *i3cdev;
 	const struct i3c_driver *i3cdrv;
+	u8 static_addr_method = 0;
 
 	if (dev->type != &i3c_device_type)
 		return 0;
 
 	i3cdev = dev_to_i3cdev(dev);
 	i3cdrv = drv_to_i3cdrv(drv);
-	if (i3c_device_match_id(i3cdev, i3cdrv->id_table))

drivers/i3c/master.c static int i3c_device_match(struct device *dev, const struct device_driver *drv) · line 365

Quoted 10 lines
drivers/i3c/master.c +10 -0
+
+	if (i3cdev->desc && i3cdev->desc->boardinfo)
+		static_addr_method = i3cdev->desc->boardinfo->static_addr_method;
+
+	/*
+	 * SETAASA-based devices need not always have a matching ID since
+	 * it is not mandatory for such devices to implement deviceinfo
+	 * CCC commands. Allow them to register through DT or ACPI.
+	 */
+	if (i3cdrv->id_table && i3c_device_match_id(i3cdev, i3cdrv->id_table))

[Severity: High]
This is a pre-existing issue, but can i3c_device_match_id() read uninitialized stack memory here?

During device unregistration, i3c_master_unregister_i3c_devs() explicitly sets i3cdev->dev->desc = NULL before calling device_unregister().

If driver matching occurs concurrently, i3c_device_match_id() might be called while desc is NULL:

drivers/i3c/device.c:i3c_device_match_id() {
    struct i3c_device_info devinfo;
    ...
    i3c_device_get_info(i3cdev, &devinfo);
    manuf = I3C_PID_MANUF_ID(devinfo.pid);
    ...
}

Because i3c_device_get_info() does not initialize devinfo when desc is NULL, this leaves stack memory to be read unconditionally and used in the matching logic. This same uninitialized memory can also leak to userspace through the uevent MODALIAS string in i3c_device_uevent() when desc is NULL.

Should i3c_device_match_id() be guarded by checking i3cdev->desc, similar to how static_addr_method is guarded above?

drivers/i3c/master.c static int i3c_device_match(struct device *dev, const struct device_driver *drv) · line 376

Quoted 11 lines
drivers/i3c/master.c +7 -0
 		return 1;
 
+	if (static_addr_method & I3C_ADDR_METHOD_SETAASA) {
+		if (of_driver_match_device(dev, drv))
+			return 1;
+		if (acpi_driver_match_device(dev, drv))
+			return 1;
+	}
+
 	return 0;
 }

Also, while reviewing this, I noticed another pre-existing issue during device unregistration:

[Severity: High]
This isn't a bug introduced by this patch, but can bcr_show() and dcr_show() dereference a NULL pointer?

When i3c_master_unregister_i3c_devs() clears i3cdev->dev->desc = NULL, there is a race window before the sysfs files are removed in device_del(). If a user reads the bcr or dcr sysfs attributes during this window, dev_to_i3cdesc() will return NULL:

drivers/i3c/master.c:bcr_show() {
    ...
    desc = dev_to_i3cdesc(dev);
    ret = sysfs_emit(buf, "0x%02x\n", desc->info.bcr);
    ...
}

Could this lead to a NULL pointer dereference?

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=6
author
#15 lore raw

PATCH v7 7/12 i3c: dw-i3c-master: Add SETAASA as supported CCC

In reply to Akhil R

Add SETAASA and SETHID to the supported list of CCC commands for DesignWare I3C host controller.

SETAASA is a broadcast command that assigns predefined static addresses to all I3C devices on the bus.

SETHID is to stop HID bit flipping by the SPD Hub to which the SPD devices are connected. It is a prerequisite command to be sent before SETAASA as recommended by JESD300-5 and JESD403 sideband bus specifications.

---
 drivers/i3c/master/dw-i3c-master.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/i3c/master/dw-i3c-master.c +2 -0
index 2f8c0c4683e0..29030fd9594a 100644
@@ -309,6 +309,8 @@ static bool dw_i3c_master_supports_ccc_cmd(struct i3c_master_controller *m,
 	case I3C_CCC_GETSTATUS:
 	case I3C_CCC_GETMXDS:
 	case I3C_CCC_GETHDRCAP:
+	case I3C_CCC_SETAASA:
+	case I3C_CCC_VENDOR(0, true): /* SETHID */
 		return true;
 	default:
 		return false;
Signature · 1 line
2.43.0
#16 lore raw

Re: [PATCH v7 07/12] i3c: dw-i3c-master: Add SETAASA as supported CCC

Sashiko has reviewed this patch and found no issues. It looks great!
Quoted 6 lines

Add SETAASA and SETHID to the supported list of CCC commands for DesignWare I3C host controller.

SETAASA is a broadcast command that assigns predefined static addresses to all I3C devices on the bus.

Quoted 6 lines

SETHID is to stop HID bit flipping by the SPD Hub to which the SPD devices are connected. It is a prerequisite command to be sent before SETAASA as recommended by JESD300-5 and JESD403 sideband bus specifications.

Sashiko has reviewed this patch and found no issues. It looks great!

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=7
author
#17 lore raw

PATCH v7 8/12 i3c: dw-i3c-master: Add ACPI core clock frequency quirk

In reply to Akhil R

Some ACPI-enumerated devices like Tegra410 do not expose the controller core clock through the clk framework. Unlike device tree, ACPI on Arm does not model clock providers. The hardware is expected to have its clocks enabled by firmware before the OS takes over.

Make the core clock optional and allow selected ACPI devices to provide the core clock rate through the "clock-frequency" _DSD property when the core clock is absent.

Resolve device quirks before acquiring the core clock so platforms without the ACPI skip-clock quirk still fail probe immediately when the clock is missing, before any MMIO access.

---
 drivers/i3c/master/dw-i3c-master.c | 45 ++++++++++++++++++++++--------
 1 file changed, 33 insertions(+), 12 deletions(-)
drivers/i3c/master/dw-i3c-master.c +33 -12
index 29030fd9594a..273d7a002099 100644
@@ -242,6 +242,7 @@
 /* List of quirks */
 #define AMD_I3C_OD_PP_TIMING		BIT(1)
 #define DW_I3C_DISABLE_RUNTIME_PM_QUIRK	BIT(2)
+#define DW_I3C_ACPI_SKIP_CLK_RST	BIT(3)
 
 struct dw_i3c_cmd {
 	u32 cmd_lo;
@@ -556,13 +557,28 @@ static void dw_i3c_master_set_intr_regs(struct dw_i3c_master *master)
 	writel(IBI_REQ_REJECT_ALL, master->regs + IBI_MR_REQ_REJECT);
 }
 
+static unsigned long dw_i3c_master_get_core_rate(struct dw_i3c_master *master)
+{
+	unsigned int core_rate_prop;
+
+	if (master->core_clk)
+		return clk_get_rate(master->core_clk);
+
+	if (device_property_read_u32(master->dev, "clock-frequency", &core_rate_prop)) {
+		dev_err(master->dev, "missing clock-frequency property\n");
+		return 0;
+	}
+
+	return core_rate_prop;
+}
+
 static int dw_i3c_clk_cfg(struct dw_i3c_master *master)
 {
 	unsigned long core_rate, core_period;
 	u32 scl_timing;
 	u8 hcnt, lcnt;
 
-	core_rate = clk_get_rate(master->core_clk);
+	core_rate = dw_i3c_master_get_core_rate(master);
 	if (!core_rate)
 		return -EINVAL;
 
@@ -615,7 +631,7 @@ static int dw_i2c_clk_cfg(struct dw_i3c_master *master)
 	u16 hcnt, lcnt;
 	u32 scl_timing;
 
-	core_rate = clk_get_rate(master->core_clk);
+	core_rate = dw_i3c_master_get_core_rate(master);
 	if (!core_rate)
 		return -EINVAL;
 
@@ -1573,14 +1589,28 @@ int dw_i3c_common_probe(struct dw_i3c_master *master,
 
 	master->dev = &pdev->dev;
 
+	if (has_acpi_companion(&pdev->dev)) {
+		quirks = (unsigned long)device_get_match_data(&pdev->dev);
+	} else if (pdev->dev.of_node) {
+		drvdata = device_get_match_data(&pdev->dev);
+		if (drvdata)
+			quirks = drvdata->flags;
+	}
+	master->quirks = quirks;
+
 	master->regs = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(master->regs))
 		return PTR_ERR(master->regs);
 
-	master->core_clk = devm_clk_get_enabled(&pdev->dev, NULL);
+	master->core_clk = devm_clk_get_optional_enabled(&pdev->dev, NULL);
 	if (IS_ERR(master->core_clk))
 		return PTR_ERR(master->core_clk);
 
+	if (!master->core_clk && !(master->quirks & DW_I3C_ACPI_SKIP_CLK_RST)) {
+		dev_err(&pdev->dev, "missing core clock\n");
+		return -EINVAL;
+	}
+
 	master->pclk = devm_clk_get_optional_enabled(&pdev->dev, "pclk");
 	if (IS_ERR(master->pclk))
 		return PTR_ERR(master->pclk);
@@ -1636,15 +1666,6 @@ int dw_i3c_common_probe(struct dw_i3c_master *master,
 		master->has_ibi_data = true;
 	writel(thld_ctrl, master->regs + QUEUE_THLD_CTRL);
 
-	if (has_acpi_companion(&pdev->dev)) {
-		quirks = (unsigned long)device_get_match_data(&pdev->dev);
-	} else if (pdev->dev.of_node) {
-		drvdata = device_get_match_data(&pdev->dev);
-		if (drvdata)
-			quirks = drvdata->flags;
-	}
-	master->quirks = quirks;
-
 	/* Keep controller enabled by preventing runtime suspend */
 	if (master->quirks & DW_I3C_DISABLE_RUNTIME_PM_QUIRK)
 		pm_runtime_get_noresume(&pdev->dev);
Signature · 1 line
2.43.0
#18 lore raw

Re: [PATCH v7 08/12] i3c: dw-i3c-master: Add ACPI core clock frequency quirk

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:

New issues:
- [Medium] Dead Code / Missing Feature Implementation for DW_I3C_ACPI_SKIP_CLK_RST

Pre-existing issues:
- [High] Runtime PM and Resource Cleanup Bug during driver unbind
- [High] Type Confusion and Pointer Truncation for PRP0001 ACPI Devices --

--- Patch [8]: [PATCH v7 08/12] i3c: dw-i3c-master: Add ACPI core clock frequency quirk --- commit 7469594594011ab29fdf220e06f76b074bff94d1
Author: Akhil R <[email protected]>

i3c: dw-i3c-master: Add ACPI core clock frequency quirk

Some ACPI-enumerated devices like Tegra410 do not expose the controller core clock through the clk framework. Unlike device tree, ACPI on Arm does not model clock providers. The hardware is expected to have its clocks enabled by firmware before the OS takes over.

Quoted 4 lines
drivers/i3c/master/dw-i3c-master.c +0 -0
index 29030fd9594aa..273d7a0020993 100644

[ ... ]

drivers/i3c/master/dw-i3c-master.c int dw_i3c_common_probe(struct dw_i3c_master *master, · line 1593

Quoted 6 lines
drivers/i3c/master/dw-i3c-master.c +2 -0
@@ -1573,14 +1589,28 @@ int dw_i3c_common_probe(struct dw_i3c_master *master,
 
 	master->dev = &pdev->dev;
 
+	if (has_acpi_companion(&pdev->dev)) {
+		quirks = (unsigned long)device_get_match_data(&pdev->dev);

[Severity: High]
This isn't a bug introduced by this patch, but does this cast cause type confusion for PRP0001 ACPI devices?

When an ACPI device binds to the driver using the PRP0001 OF-compatibility mechanism, device_get_match_data() delegates to the OF match table and returns a pointer to struct dw_i3c_drvdata.

Because has_acpi_companion() returns true, this branch will cast that pointer directly to an unsigned long, treating the memory address as a bitmask of quirks. This could corrupt the driver state.

Quoted 7 lines
drivers/i3c/master/dw-i3c-master.c +7 -0
+	} else if (pdev->dev.of_node) {
+		drvdata = device_get_match_data(&pdev->dev);
+		if (drvdata)
+			quirks = drvdata->flags;
+	}
+	master->quirks = quirks;
+

drivers/i3c/master/dw-i3c-master.c int dw_i3c_common_probe(struct dw_i3c_master *master, · line 1609

Quoted 10 lines
drivers/i3c/master/dw-i3c-master.c +2 -1
 	master->regs = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(master->regs))
 		return PTR_ERR(master->regs);
 
-	master->core_clk = devm_clk_get_enabled(&pdev->dev, NULL);
+	master->core_clk = devm_clk_get_optional_enabled(&pdev->dev, NULL);
 	if (IS_ERR(master->core_clk))
 		return PTR_ERR(master->core_clk);
 
+	if (!master->core_clk && !(master->quirks & DW_I3C_ACPI_SKIP_CLK_RST)) {

[Severity: Medium]
Will this driver still fail to probe on ACPI platforms without a core clock?

The code checks if DW_I3C_ACPI_SKIP_CLK_RST is set in master->quirks. However, it appears this quirk is never included in the driver's ACPI match table (amd_i3c_device_match). Without being populated by the ACPI matching framework, the feature seems to act as dead code.

drivers/i3c/master/dw-i3c-master.c int dw_i3c_common_probe(struct dw_i3c_master *master, · line 1613

Quoted 4 lines
drivers/i3c/master/dw-i3c-master.c +4 -0
+		dev_err(&pdev->dev, "missing core clock\n");
+		return -EINVAL;
+	}
+

[ ... ]

[Severity: High]
This isn't a bug introduced by this patch, but is there a runtime PM and resource cleanup issue during driver unbind?

When the device is removed (e.g., driver unbound via sysfs) while in a runtime-suspended state, or an autosuspend timer fires concurrently with the remove callback, the driver core calls pm_runtime_put_sync() before remove.

This means the device may be suspended when dw_i3c_common_remove() executes. The callback unregisters the I3C master:

dw_i3c_common_remove()

    i3c_master_unregister(&master->base)
        dw_i3c_master_bus_cleanup()
            dw_i3c_master_disable()

This access to the DEVICE_CTRL MMIO register happens while the controller might be unclocked, which could cause a synchronous external abort (kernel panic).

Would it be safer to wake the device via pm_runtime_get_sync() before unregistering the bus?

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=8
author
#19 lore raw

PATCH v7 9/12 i3c: dw-i3c-master: Add ACPI ID for Tegra410

In reply to Akhil R

Update variable names to generic names and add Tegra410 ACPI ID to support the I3C controller in Tegra410, which is a DesignWare I3C host controller.

---
 drivers/i3c/master/dw-i3c-master.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/i3c/master/dw-i3c-master.c +4 -3
index 273d7a002099..12069aa56fad 100644
@@ -1855,11 +1855,12 @@ static const struct of_device_id dw_i3c_master_of_match[] = {
 };
 MODULE_DEVICE_TABLE(of, dw_i3c_master_of_match);
 
-static const struct acpi_device_id amd_i3c_device_match[] = {
+static const struct acpi_device_id dw_i3c_master_acpi_match[] = {
 	{ "AMDI0015", AMD_I3C_OD_PP_TIMING },
+	{ "NVDA2018", DW_I3C_ACPI_SKIP_CLK_RST },
 	{ }
 };
-MODULE_DEVICE_TABLE(acpi, amd_i3c_device_match);
+MODULE_DEVICE_TABLE(acpi, dw_i3c_master_acpi_match);
 
 static struct platform_driver dw_i3c_driver = {
 	.probe = dw_i3c_probe,
@@ -1868,7 +1869,7 @@ static struct platform_driver dw_i3c_driver = {
 	.driver = {
 		.name = "dw-i3c-master",
 		.of_match_table = dw_i3c_master_of_match,
-		.acpi_match_table = amd_i3c_device_match,
+		.acpi_match_table = dw_i3c_master_acpi_match,
 		.pm = &dw_i3c_pm_ops,
 	},
 };
Signature · 1 line
2.43.0
#20 lore raw

Re: [PATCH v7 09/12] i3c: dw-i3c-master: Add ACPI ID for Tegra410

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] Pointer address assigned to an integer bitmask variable for quirks under ACPI PRP0001 matching. --

--- Patch [9]: [PATCH v7 09/12] i3c: dw-i3c-master: Add ACPI ID for Tegra410 --- commit 4b2659ca2047946e238179b5da94407cbd49ffa2
Author: Akhil R <[email protected]>

i3c: dw-i3c-master: Add ACPI ID for Tegra410

Update variable names to generic names and add Tegra410 ACPI ID to support the I3C controller in Tegra410, which is a DesignWare I3C host controller.

Quoted 5 lines
drivers/i3c/master/dw-i3c-master.c +0 -0
index 273d7a0020993..12069aa56fad8 100644

drivers/i3c/master/dw-i3c-master.c static const struct of_device_id dw_i3c_master_of_match[] = { · line 1862

Quoted 9 lines
drivers/i3c/master/dw-i3c-master.c +2 -1
@@ -1855,11 +1855,12 @@ static const struct of_device_id dw_i3c_master_of_match[] = {
 };
 MODULE_DEVICE_TABLE(of, dw_i3c_master_of_match);
 
-static const struct acpi_device_id amd_i3c_device_match[] = {
+static const struct acpi_device_id dw_i3c_master_acpi_match[] = {
 	{ "AMDI0015", AMD_I3C_OD_PP_TIMING },
+	{ "NVDA2018", DW_I3C_ACPI_SKIP_CLK_RST },
 	{ }
 };

[Severity: High]
This is a pre-existing issue, but does dw_i3c_common_probe() improperly cast pointers to integer bitmasks under ACPI PRP0001 matching?

If a system uses the ACPI PRP0001 HID alongside the "altr,agilex5-dw-i3c-master" compatible string, device_get_match_data() will return a pointer to altr_agilex5_drvdata.

Because the device has an ACPI companion, the probe function processes it like this:

drivers/i3c/master/dw-i3c-master.c:dw_i3c_common_probe() {
    ...
	if (has_acpi_companion(&pdev->dev)) {
		quirks = (unsigned long)device_get_match_data(&pdev->dev);
	} else if (pdev->dev.of_node) {
    ...
}

Will this cause the memory address of altr_agilex5_drvdata to be cast directly to an integer bitmask, resulting in random quirks being activated based on memory alignment (e.g., falsely applying DW_I3C_ACPI_SKIP_CLK_RST or DW_I3C_DISABLE_RUNTIME_PM_QUIRK)?

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=9
author
#21 lore raw

PATCH v7 10/12 hwmon: spd5118: Remove 16-bit addressing

In reply to Akhil R

The intent of introducing 16-bit addressing was to support I3C, but it turns out that I3C does not require reading the Legacy Mode register, nor any specific encoding for page translation. The testing of 16-bit code was limited and there are no known users for this feature. Remove the sections that support 16-bit addressing and prepare the driver to support I3C appropriately.

---
 drivers/hwmon/spd5118.c | 79 +++--------------------------------------
 1 file changed, 5 insertions(+), 74 deletions(-)
drivers/hwmon/spd5118.c +5 -74
index cc40661cab21..6ba37a719300 100644
@@ -66,9 +66,6 @@ static const unsigned short normal_i2c[] = {
 #define SPD5118_EEPROM_BASE		0x80
 #define SPD5118_EEPROM_SIZE		(SPD5118_PAGE_SIZE * SPD5118_NUM_PAGES)
 
-#define PAGE_ADDR0(page)		(((page) & BIT(0)) << 6)
-#define PAGE_ADDR1_4(page)		(((page) & GENMASK(4, 1)) >> 1)
-
 /* Temperature unit in millicelsius */
 #define SPD5118_TEMP_UNIT		(MILLIDEGREE_PER_DEGREE / 4)
 /* Representable temperature range in millicelsius */
@@ -78,7 +75,6 @@ static const unsigned short normal_i2c[] = {
 struct spd5118_data {
 	struct regmap *regmap;
 	struct mutex nvmem_lock;
-	bool is_16bit;
 };
 
 /* hwmon */
@@ -348,12 +344,7 @@ static ssize_t spd5118_nvmem_read_page(struct spd5118_data *data, char *buf,
 	if (offset + count > SPD5118_PAGE_SIZE)
 		count = SPD5118_PAGE_SIZE - offset;
 
-	if (data->is_16bit) {
-		addr = SPD5118_EEPROM_BASE | PAGE_ADDR0(page) |
-		  (PAGE_ADDR1_4(page) << 8);
-	} else {
-		addr = page * 0x100 + SPD5118_EEPROM_BASE;
-	}
+	addr = page * 0x100 + SPD5118_EEPROM_BASE;
 	err = regmap_bulk_read(regmap, addr + offset, buf, count);
 	if (err)
 		return err;
@@ -473,15 +464,6 @@ static const struct regmap_config spd5118_regmap8_config = {
 	.num_ranges = ARRAY_SIZE(spd5118_i2c_regmap_range_cfg),
 };
 
-static const struct regmap_config spd5118_regmap16_config = {
-	.reg_bits = 16,
-	.val_bits = 8,
-	.max_register = 0x7ff,
-	.writeable_reg = spd5118_writeable_reg,
-	.volatile_reg = spd5118_volatile_reg,
-	.cache_type = REGCACHE_MAPLE,
-};
-
 static int spd5118_suspend(struct device *dev)
 {
 	struct spd5118_data *data = dev_get_drvdata(dev);
@@ -519,8 +501,7 @@ static int spd5118_resume(struct device *dev)
 
 static DEFINE_SIMPLE_DEV_PM_OPS(spd5118_pm_ops, spd5118_suspend, spd5118_resume);
 
-static int spd5118_common_probe(struct device *dev, struct regmap *regmap,
-				bool is_16bit)
+static int spd5118_common_probe(struct device *dev, struct regmap *regmap)
 {
 	unsigned int capability, revision, vendor, bank;
 	struct spd5118_data *data;
@@ -537,8 +518,6 @@ static int spd5118_common_probe(struct device *dev, struct regmap *regmap,
 	if (!(capability & SPD5118_CAP_TS_SUPPORT))
 		return -ENODEV;
 
-	data->is_16bit = is_16bit;
-
 	err = regmap_read(regmap, SPD5118_REG_REVISION, &revision);
 	if (err)
 		return err;
@@ -680,69 +659,21 @@ static int spd5118_i2c_init(struct i2c_client *client)
 	return 0;
 }
 
-/*
- * 16-bit addressing note:
- *
- * If I2C_FUNC_I2C is not supported by an I2C adapter driver, regmap uses
- * SMBus operations as alternative. To simulate a read operation with a 16-bit
- * address, it writes the address using i2c_smbus_write_byte_data(), followed
- * by one or more calls to i2c_smbus_read_byte() to read the data.
- * Per spd5118 standard, a read operation after writing the address must start
- * with <Sr> (Repeat Start). However, a SMBus read byte operation starts with
- * <S> (Start). This resets the register address in the spd5118 chip. As result,
- * i2c_smbus_read_byte() always returns data from register address 0x00.
- *
- * A working alternative to access chips with 16-bit register addresses in the
- * absence of I2C_FUNC_I2C support is not known.
- *
- * For this reason, 16-bit addressing can only be supported with I2C if the
- * adapter supports I2C_FUNC_I2C.
- *
- * For I2C, the addressing mode selected by the BIOS must not be changed.
- * Experiments show that at least some PC BIOS versions will not change the
- * addressing mode on a soft reboot and end up in setup, claiming that some
- * configuration change happened. This will happen again after a power cycle,
- * which does reset the addressing mode. To prevent this from happening,
- * detect if 16-bit addressing is enabled and always use the currently
- * configured addressing mode.
- */
-
 static int spd5118_i2c_probe(struct i2c_client *client)
 {
-	const struct regmap_config *config;
 	struct device *dev = &client->dev;
 	struct regmap *regmap;
-	int err, mode;
-	bool is_16bit;
+	int err;
 
 	err = spd5118_i2c_init(client);
 	if (err)
 		return err;
 
-	mode = i2c_smbus_read_byte_data(client, SPD5118_REG_I2C_LEGACY_MODE);
-	if (mode < 0)
-		return mode;
-
-	is_16bit = mode & SPD5118_LEGACY_MODE_ADDR;
-	if (is_16bit) {
-		/*
-		 * See 16-bit addressing note above explaining why it is
-		 * necessary to check for I2C_FUNC_I2C support here.
-		 */
-		if (!i2c_check_functionality(client->adapter, I2C_FUNC_I2C)) {
-			dev_err(dev, "Adapter does not support 16-bit register addresses\n");
-			return -ENODEV;
-		}
-		config = &spd5118_regmap16_config;
-	} else {
-		config = &spd5118_regmap8_config;
-	}
-
-	regmap = devm_regmap_init_i2c(client, config);
+	regmap = devm_regmap_init_i2c(client, &spd5118_regmap8_config);
 	if (IS_ERR(regmap))
 		return dev_err_probe(dev, PTR_ERR(regmap), "regmap init failed\n");
 
-	return spd5118_common_probe(dev, regmap, is_16bit);
+	return spd5118_common_probe(dev, regmap);
 }
 
 static const struct i2c_device_id spd5118_i2c_id[] = {
Signature · 1 line
2.43.0
#22 lore raw

Re: [PATCH v7 10/12] hwmon: spd5118: Remove 16-bit addressing

Sashiko has reviewed this patch and found no issues. It looks great!
Quoted 10 lines

The intent of introducing 16-bit addressing was to support I3C, but it turns out that I3C does not require reading the Legacy Mode register, nor any specific encoding for page translation. The testing of 16-bit code was limited and there are no known users for this feature. Remove the sections that support 16-bit addressing and prepare the driver to support I3C appropriately.

Sashiko has reviewed this patch and found no issues. It looks great!

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=10
author
#23 lore raw

PATCH v7 11/12 hwmon: spd5118: Add I3C support

In reply to Akhil R

Add a regmap config and a probe function to support I3C-based communication with SPD5118 devices.

On an I3C bus, SPD5118 devices are enumerated via SETAASA and always require an ACPI or device tree entry. Device matching is hence through the OF match tables only and does not need an I3C class match table. The device identity is verified in the type registers before proceeding to the common probe function.

---
 drivers/hwmon/Kconfig   |  9 ++++---
 drivers/hwmon/spd5118.c | 56 ++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 61 insertions(+), 4 deletions(-)
drivers/hwmon/Kconfig +6 -3
index b2e445b5536e..d9feba036ba4 100644
@@ -2370,12 +2370,15 @@ config SENSORS_INA3221
 
 config SENSORS_SPD5118
 	tristate "SPD5118 Compliant Temperature Sensors"
-	depends on I2C
+	depends on I3C_OR_I2C
 	select REGMAP_I2C
+	select REGMAP_I3C if I3C
 	help
 	  If you say yes here you get support for SPD5118 (JEDEC JESD300)
-	  compliant temperature sensors. Such sensors are found on DDR5 memory
-	  modules.
+	  compliant temperature sensors using I2C or I3C bus interface.
+	  Such sensors are found on DDR5 memory modules.
+
+	  This driver supports both I2C and I3C interfaces.
 
 	  This driver can also be built as a module. If so, the module
 	  will be called spd5118.
drivers/hwmon/spd5118.c +55 -1

74 lines not on this page · raw message

Signature · 1 line
2.43.0
#24 lore raw

Re: [PATCH v7 11/12] hwmon: spd5118: Add I3C support

Sashiko has reviewed this patch and found no issues. It looks great!
Quoted 11 lines

Add a regmap config and a probe function to support I3C-based communication with SPD5118 devices.

On an I3C bus, SPD5118 devices are enumerated via SETAASA and always require an ACPI or device tree entry. Device matching is hence through the OF match tables only and does not need an I3C class match table. The device identity is verified in the type registers before proceeding to the common probe function.

Sashiko has reviewed this patch and found no issues. It looks great!

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=11
author
#25 lore raw

PATCH v7 12/12 arm64: defconfig: Enable I3C and SPD5118 hwmon

In reply to Akhil R

Enable I3C subsystem (I3C), Synopsys DesignWare I3C master controller (DW_I3C_MASTER), and SPD5118 hwmon temperature sensor (SENSORS_SPD5118) as modules.

The NVIDIA Vera CPU uses SOCAMM LPDDR5X memory module, which contains SPD5118 (JEDEC JESD300) compliant temperature sensor. This sensor is accessible over the I3C bus through the DesignWare I3C controller present on the SoC. Enabling these configs allows monitoring memory module temperatures on platforms such as Vera Rubin. Vera is an ACPI-based platform and does not use device tree.

---
 arch/arm64/configs/defconfig | 3 +++
 1 file changed, 3 insertions(+)
arch/arm64/configs/defconfig +3 -0
index c267f0906460..36a7cafef3a0 100644
@@ -639,6 +639,8 @@ CONFIG_I2C_UNIPHIER_F=y
 CONFIG_I2C_XILINX=m
 CONFIG_I2C_RCAR=y
 CONFIG_I2C_CROS_EC_TUNNEL=y
+CONFIG_I3C=m
+CONFIG_DW_I3C_MASTER=m
 CONFIG_SPI=y
 CONFIG_SPI_APPLE=m
 CONFIG_SPI_ARMADA_3700=y
@@ -768,6 +770,7 @@ CONFIG_SENSORS_SL28CPLD=m
 CONFIG_SENSORS_AMC6821=m
 CONFIG_SENSORS_INA2XX=m
 CONFIG_SENSORS_INA3221=m
+CONFIG_SENSORS_SPD5118=m
 CONFIG_SENSORS_TMP102=m
 CONFIG_THERMAL_GOV_POWER_ALLOCATOR=y
 CONFIG_CPU_THERMAL=y
Signature · 1 line
2.43.0
#26 lore raw

Re: [PATCH v7 12/12] arm64: defconfig: Enable I3C and SPD5118 hwmon

Sashiko has reviewed this patch and found no issues. It looks great!
Quoted 4 lines

Enable I3C subsystem (I3C), Synopsys DesignWare I3C master controller (DW_I3C_MASTER), and SPD5118 hwmon temperature sensor (SENSORS_SPD5118) as modules.

Quoted 8 lines

The NVIDIA Vera CPU uses SOCAMM LPDDR5X memory module, which contains SPD5118 (JEDEC JESD300) compliant temperature sensor. This sensor is accessible over the I3C bus through the DesignWare I3C controller present on the SoC. Enabling these configs allows monitoring memory module temperatures on platforms such as Vera Rubin. Vera is an ACPI-based platform and does not use device tree.

Sashiko has reviewed this patch and found no issues. It looks great!

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=12
maintainer Applied
#27 lore raw

Re: (subset) [PATCH v7 00/12] Support ACPI and SETAASA device discovery

In reply to Akhil R
Quoted 5 lines · Akhil R

This patch series adds SETAASA device discovery to the I3C subsystem, enabling support for SPD5118 temperature sensors found on DDR5 memory modules. The changes also add ACPI support for all existing DAA methods like SETDASA, SETNEWDA as well as I2C devices on I3C bus.

Quoted 7 lines

SPD5118 and similar devices on DDR5 memory modules differ from typical I3C devices in their initialization. They use SETAASA broadcast CCC instead of ENTDAA for address assignment, and per JEDEC specification, are not required to have a Provisioned ID or implement standard device information CCC commands (GETPID, GETDCR, GETBCR).

[...]

Applied, thanks!

[01/12] dt-bindings: i3c: Add mipi-i3c-static-method to support SETAASA

        https://git.kernel.org/i3c/c/29593f94a91c
[02/12] i3c: master: Use unified device property interface
        https://git.kernel.org/i3c/c/f1e00dcda80f
[03/12] i3c: master: Support ACPI enumeration of child devices
        https://git.kernel.org/i3c/c/c5cc24aebe83
[04/12] i3c: master: Add support for devices using SETAASA
        https://git.kernel.org/i3c/c/46276c855815
[05/12] i3c: master: Add support for devices without PID
        https://git.kernel.org/i3c/c/9885e8c54210
[06/12] i3c: master: match I3C device through DT and ACPI
        https://git.kernel.org/i3c/c/95bc66704fd5
[07/12] i3c: dw-i3c-master: Add SETAASA as supported CCC
        https://git.kernel.org/i3c/c/d3aa6f594db2
[08/12] i3c: dw-i3c-master: Add ACPI core clock frequency quirk
        https://git.kernel.org/i3c/c/581cca59b5fa
[09/12] i3c: dw-i3c-master: Add ACPI ID for Tegra410
        https://git.kernel.org/i3c/c/77de3d43f74d
[10/12] hwmon: spd5118: Remove 16-bit addressing
        https://git.kernel.org/i3c/c/89eee0b30226
[11/12] hwmon: spd5118: Add I3C support
        https://git.kernel.org/i3c/c/f76be53203bb

Best regards,

Signature · 3 lines
Alexandre Belloni, co-owner and COO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com