[PATCH v5 0/6] Introduce PRU remoteproc consumer API

[PATCH v5 0/6] Introduce PRU remoteproc consumer API

[Puranjay Mohan]

This is the v5 of the patch series [1]. The 6th patch in the series was
added in v4 and it had some checkpatch.pl errors which are removed here.

I have posted two more patch series that depend on this series, one to
the soc tree and another to the networking tree. I had sent all the 3
series, including this one as RFC [2] to get comments and to explain the
dependencies.

The Programmable Real-Time Unit and Industrial Communication Subsystem
(PRU-ICSS or simply PRUSS) on various TI SoCs consists of dual 32-bit
RISC cores (Programmable Real-Time Units, or PRUs) for program execution.

There are 3 foundation components for PRUSS subsystem: the PRUSS platform
driver, the PRUSS INTC driver and the PRUSS remoteproc driver. All were
already merged and can be found under:
1) drivers/soc/ti/pruss.c
   Documentation/devicetree/bindings/soc/ti/ti,pruss.yaml
2) drivers/irqchip/irq-pruss-intc.c
   Documentation/devicetree/bindings/interrupt-controller/ti,pruss-intc.yaml
3) drivers/remoteproc/pru_rproc.c
   Documentation/devicetree/bindings/remoteproc/ti,pru-rproc.yaml

The programmable nature of the PRUs provide flexibility to implement custom
peripheral interfaces, fast real-time responses, or specialized data handling.
Example of a PRU consumer drivers will be:
  - Software UART over PRUSS
  - PRU-ICSS Ethernet EMAC

In order to make usage of common PRU resources and allow the consumer drivers to
configure the PRU hardware for specific usage the PRU API is introduced.

[1] https://patchwork.kernel.org/project/linux-remoteproc/cover/20220603121520.13730-1-p-mohan@ti.com/
[2] https://patchwork.kernel.org/project/linux-remoteproc/cover/20220406094358.7895-1-p-mohan@ti.com/

Thanks and Regards,
Puranjay Mohan

Roger Quadros (1):
  remoteproc: pru: Add pru_rproc_set_ctable() function

Suman Anna (2):
  dt-bindings: remoteproc: Add PRU consumer bindings
  remoteproc: pru: Make sysfs entries read-only for PRU client driven
    boots

Tero Kristo (3):
  remoteproc: pru: Add APIs to get and put the PRU cores
  remoteproc: pru: Configure firmware based on client setup
  remoteproc: pru: add support for configuring GPMUX based on client
    setup

 .../bindings/remoteproc/ti,pru-consumer.yaml  |  69 +++++
 drivers/remoteproc/pru_rproc.c                | 254 +++++++++++++++++-
 include/linux/pruss.h                         |  78 ++++++
 3 files changed, 396 insertions(+), 5 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/remoteproc/ti,pru-consumer.yaml
 create mode 100644 include/linux/pruss.h

-- 
2.17.1

[PATCH v5 1/6] dt-bindings: remoteproc: Add PRU consumer bindings

[Puranjay Mohan]

From: Suman Anna <s-anna@ti.com>

Add a YAML binding document for PRU consumers. The binding includes
all the common properties that can be used by different PRU consumer
or application nodes and supported by the PRU remoteproc driver.
These are used to configure the PRU hardware for specific user
applications.

The application nodes themselves should define their own bindings.

Co-developed-by: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Suman Anna <s-anna@ti.com>
Co-developed-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
Signed-off-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
Reviewed-by: Rob Herring <robh@kernel.org>
---
V3->V4:
* Addressed Rob's comments regarding max and min Items.
* removed the dependencies tag as it was redundant.
---
 .../bindings/remoteproc/ti,pru-consumer.yaml  | 69 +++++++++++++++++++
 1 file changed, 69 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/remoteproc/ti,pru-consumer.yaml

diff --git a/Documentation/devicetree/bindings/remoteproc/ti,pru-consumer.yaml b/Documentation/devicetree/bindings/remoteproc/ti,pru-consumer.yaml
new file mode 100644
index 000000000000..df384b44259b
--- /dev/null
+++ b/Documentation/devicetree/bindings/remoteproc/ti,pru-consumer.yaml
@@ -0,0 +1,69 @@
+# SPDX-License-Identifier: (GPL-2.0-only or BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/remoteproc/ti,pru-consumer.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Common TI PRU Consumer Binding
+
+maintainers:
+  - Suman Anna <s-anna@ti.com>
+
+description: |
+  A PRU application/consumer/user node typically uses one or more PRU device
+  nodes to implement a PRU application/functionality. Each application/client
+  node would need a reference to at least a PRU node, and optionally define
+  some properties needed for hardware/firmware configuration. The below
+  properties are a list of common properties supported by the PRU remoteproc
+  infrastructure.
+
+  The application nodes shall define their own bindings like regular platform
+  devices, so below are in addition to each node's bindings.
+
+properties:
+  ti,prus:
+    $ref: /schemas/types.yaml#/definitions/phandle-array
+    description: phandles to the PRU, RTU or Tx_PRU nodes used
+    minItems: 1
+    maxItems: 6
+    items:
+      maxItems: 1
+
+  firmware-name:
+    $ref: /schemas/types.yaml#/definitions/string-array
+    minItems: 1
+    maxItems: 6
+    description: |
+      firmwares for the PRU cores, the default firmware for the core from
+      the PRU node will be used if not provided. The firmware names should
+      correspond to the PRU cores listed in the 'ti,prus' property
+
+  ti,pruss-gp-mux-sel:
+    $ref: /schemas/types.yaml#/definitions/uint32-array
+    minItems: 1
+    maxItems: 6
+    items:
+      enum: [0, 1, 2, 3, 4]
+    description: |
+      array of values for the GP_MUX_SEL under PRUSS_GPCFG register for a PRU.
+      This selects the internal muxing scheme for the PRU instance. Values
+      should correspond to the PRU cores listed in the 'ti,prus' property. The
+      GP_MUX_SEL setting is a per-slice setting (one setting for PRU0, RTU0,
+      and Tx_PRU0 on K3 SoCs). Use the same value for all cores within the
+      same slice in the associative array. If the array size is smaller than
+      the size of 'ti,prus' property, the default out-of-reset value (0) for the
+      PRU core is used.
+
+required:
+  - ti,prus
+
+additionalProperties: true
+
+examples:
+  - |
+    /* PRU application node example */
+    pru-app {
+        ti,prus = <&pru0>, <&pru1>;
+        firmware-name = "pruss-app-fw0", "pruss-app-fw1";
+        ti,pruss-gp-mux-sel = <2>, <1>;
+    };
-- 
2.17.1

Re: [PATCH v5 1/6] dt-bindings: remoteproc: Add PRU consumer bindings

[Mathieu Poirier]

On Tue, Jun 07, 2022 at 10:26:45AM +0530, Puranjay Mohan wrote:
> From: Suman Anna <s-anna@ti.com>
> 
> Add a YAML binding document for PRU consumers. The binding includes
> all the common properties that can be used by different PRU consumer
> or application nodes and supported by the PRU remoteproc driver.
> These are used to configure the PRU hardware for specific user
> applications.
> 
> The application nodes themselves should define their own bindings.
> 
> Co-developed-by: Tero Kristo <t-kristo@ti.com>
> Signed-off-by: Tero Kristo <t-kristo@ti.com>
> Signed-off-by: Suman Anna <s-anna@ti.com>
> Co-developed-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
> Signed-off-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
> Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
> Reviewed-by: Rob Herring <robh@kernel.org>
> ---
> V3->V4:
> * Addressed Rob's comments regarding max and min Items.
> * removed the dependencies tag as it was redundant.
> ---
>  .../bindings/remoteproc/ti,pru-consumer.yaml  | 69 +++++++++++++++++++
>  1 file changed, 69 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/remoteproc/ti,pru-consumer.yaml
> 
> diff --git a/Documentation/devicetree/bindings/remoteproc/ti,pru-consumer.yaml b/Documentation/devicetree/bindings/remoteproc/ti,pru-consumer.yaml
> new file mode 100644
> index 000000000000..df384b44259b
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/remoteproc/ti,pru-consumer.yaml
> @@ -0,0 +1,69 @@
> +# SPDX-License-Identifier: (GPL-2.0-only or BSD-2-Clause)
> +%YAML 1.2
> +---
> +$id: http://devicetree.org/schemas/remoteproc/ti,pru-consumer.yaml#
> +$schema: http://devicetree.org/meta-schemas/core.yaml#
> +
> +title: Common TI PRU Consumer Binding
> +
> +maintainers:
> +  - Suman Anna <s-anna@ti.com>
> +
> +description: |
> +  A PRU application/consumer/user node typically uses one or more PRU device
> +  nodes to implement a PRU application/functionality. Each application/client
> +  node would need a reference to at least a PRU node, and optionally define
> +  some properties needed for hardware/firmware configuration. The below
> +  properties are a list of common properties supported by the PRU remoteproc
> +  infrastructure.
> +
> +  The application nodes shall define their own bindings like regular platform
> +  devices, so below are in addition to each node's bindings.
> +
> +properties:
> +  ti,prus:
> +    $ref: /schemas/types.yaml#/definitions/phandle-array
> +    description: phandles to the PRU, RTU or Tx_PRU nodes used
> +    minItems: 1
> +    maxItems: 6
> +    items:
> +      maxItems: 1
> +
> +  firmware-name:
> +    $ref: /schemas/types.yaml#/definitions/string-array
> +    minItems: 1
> +    maxItems: 6
> +    description: |
> +      firmwares for the PRU cores, the default firmware for the core from
> +      the PRU node will be used if not provided. The firmware names should
> +      correspond to the PRU cores listed in the 'ti,prus' property
> +
> +  ti,pruss-gp-mux-sel:
> +    $ref: /schemas/types.yaml#/definitions/uint32-array
> +    minItems: 1
> +    maxItems: 6
> +    items:
> +      enum: [0, 1, 2, 3, 4]
> +    description: |
> +      array of values for the GP_MUX_SEL under PRUSS_GPCFG register for a PRU.
> +      This selects the internal muxing scheme for the PRU instance. Values
> +      should correspond to the PRU cores listed in the 'ti,prus' property. The
> +      GP_MUX_SEL setting is a per-slice setting (one setting for PRU0, RTU0,
> +      and Tx_PRU0 on K3 SoCs). Use the same value for all cores within the
> +      same slice in the associative array. If the array size is smaller than
> +      the size of 'ti,prus' property, the default out-of-reset value (0) for the
> +      PRU core is used.
> +
> +required:
> +  - ti,prus
> +
> +additionalProperties: true
> +
> +examples:
> +  - |
> +    /* PRU application node example */
> +    pru-app {
> +        ti,prus = <&pru0>, <&pru1>;
> +        firmware-name = "pruss-app-fw0", "pruss-app-fw1";
> +        ti,pruss-gp-mux-sel = <2>, <1>;
> +    };

It would be nice to have a full example in order to provide more context.

I am done reviewing this set.

Thanks,
Mathieu

> -- 
> 2.17.1
> 

[PATCH v5 2/6] remoteproc: pru: Add APIs to get and put the PRU cores

[Puranjay Mohan]

From: Tero Kristo <t-kristo@ti.com>

Add two new APIs, pru_rproc_get() and pru_rproc_put(), to the PRU
driver to allow client drivers to acquire and release the remoteproc
device associated with a PRU core. The PRU cores are treated as
resources with only one client owning it at a time.

The pru_rproc_get() function returns the rproc handle corresponding
to a PRU core identified by the device tree "ti,prus" property under
the client node. The pru_rproc_put() is the complementary function
to pru_rproc_get().

Co-developed-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Tero Kristo <t-kristo@ti.com>
Co-developed-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
Signed-off-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
Co-developed-by: Puranjay Mohan <p-mohan@ti.com>
Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
---
 drivers/remoteproc/pru_rproc.c | 138 +++++++++++++++++++++++++++++++--
 include/linux/pruss.h          |  56 +++++++++++++
 2 files changed, 189 insertions(+), 5 deletions(-)
 create mode 100644 include/linux/pruss.h

diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
index 1777a01fa84e..7a35b400287a 100644
--- a/drivers/remoteproc/pru_rproc.c
+++ b/drivers/remoteproc/pru_rproc.c
@@ -2,12 +2,13 @@
 /*
  * PRU-ICSS remoteproc driver for various TI SoCs
  *
- * Copyright (C) 2014-2020 Texas Instruments Incorporated - https://www.ti.com/
+ * Copyright (C) 2014-2022 Texas Instruments Incorporated - https://www.ti.com/
  *
  * Author(s):
  *	Suman Anna <s-anna@ti.com>
  *	Andrew F. Davis <afd@ti.com>
  *	Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org> for Texas Instruments
+ *	Puranjay Mohan <p-mohan@ti.com>
  */
 
 #include <linux/bitops.h>
@@ -16,6 +17,7 @@
 #include <linux/module.h>
 #include <linux/of_device.h>
 #include <linux/of_irq.h>
+#include <linux/pruss.h>
 #include <linux/pruss_driver.h>
 #include <linux/remoteproc.h>
 
@@ -111,6 +113,8 @@ struct pru_private_data {
  * @rproc: remoteproc pointer for this PRU core
  * @data: PRU core specific data
  * @mem_regions: data for each of the PRU memory regions
+ * @client_np: client device node
+ * @lock: mutex to protect client usage
  * @fw_name: name of firmware image used during loading
  * @mapped_irq: virtual interrupt numbers of created fw specific mapping
  * @pru_interrupt_map: pointer to interrupt mapping description (firmware)
@@ -126,6 +130,8 @@ struct pru_rproc {
 	struct rproc *rproc;
 	const struct pru_private_data *data;
 	struct pruss_mem_region mem_regions[PRU_IOMEM_MAX];
+	struct device_node *client_np;
+	struct mutex lock; /* client access lock */
 	const char *fw_name;
 	unsigned int *mapped_irq;
 	struct pru_irq_rsc *pru_interrupt_map;
@@ -146,6 +152,125 @@ void pru_control_write_reg(struct pru_rproc *pru, unsigned int reg, u32 val)
 	writel_relaxed(val, pru->mem_regions[PRU_IOMEM_CTRL].va + reg);
 }
 
+static struct rproc *__pru_rproc_get(struct device_node *np, int index)
+{
+	struct device_node *rproc_np = NULL;
+	struct platform_device *pdev;
+	struct rproc *rproc;
+
+	rproc_np = of_parse_phandle(np, "ti,prus", index);
+	if (!rproc_np || !of_device_is_available(rproc_np))
+		return ERR_PTR(-ENODEV);
+
+	pdev = of_find_device_by_node(rproc_np);
+	of_node_put(rproc_np);
+
+	if (!pdev || !(&pdev->dev) || !((&pdev->dev)->driver))
+		/* probably PRU not yet probed */
+		return ERR_PTR(-EPROBE_DEFER);
+
+	/* make sure it is PRU rproc */
+	if (!is_pru_rproc(&pdev->dev)) {
+		put_device(&pdev->dev);
+		return ERR_PTR(-ENODEV);
+	}
+
+	rproc = platform_get_drvdata(pdev);
+	put_device(&pdev->dev);
+	if (!rproc)
+		return ERR_PTR(-EPROBE_DEFER);
+
+	get_device(&rproc->dev);
+
+	return rproc;
+}
+
+/**
+ * pru_rproc_get() - get the PRU rproc instance from a device node
+ * @np: the user/client device node
+ * @index: index to use for the ti,prus property
+ * @pru_id: optional pointer to return the PRU remoteproc processor id
+ *
+ * This function looks through a client device node's "ti,prus" property at
+ * index @index and returns the rproc handle for a valid PRU remote processor if
+ * found. The function allows only one user to own the PRU rproc resource at a
+ * time. Caller must call pru_rproc_put() when done with using the rproc, not
+ * required if the function returns a failure.
+ *
+ * When optional @pru_id pointer is passed the PRU remoteproc processor id is
+ * returned.
+ *
+ * Return: rproc handle on success, and an ERR_PTR on failure using one
+ * of the following error values
+ *    -ENODEV if device is not found
+ *    -EBUSY if PRU is already acquired by anyone
+ *    -EPROBE_DEFER is PRU device is not probed yet
+ */
+struct rproc *pru_rproc_get(struct device_node *np, int index,
+			    enum pruss_pru_id *pru_id)
+{
+	struct rproc *rproc;
+	struct pru_rproc *pru;
+	struct device *dev;
+
+	try_module_get(THIS_MODULE);
+
+	rproc = __pru_rproc_get(np, index);
+	if (IS_ERR(rproc))
+		return rproc;
+
+	pru = rproc->priv;
+	dev = &rproc->dev;
+
+	mutex_lock(&pru->lock);
+
+	if (pru->client_np) {
+		mutex_unlock(&pru->lock);
+		put_device(dev);
+		return ERR_PTR(-EBUSY);
+	}
+
+	pru->client_np = np;
+
+	mutex_unlock(&pru->lock);
+
+	if (pru_id)
+		*pru_id = pru->id;
+
+	return rproc;
+}
+EXPORT_SYMBOL_GPL(pru_rproc_get);
+
+/**
+ * pru_rproc_put() - release the PRU rproc resource
+ * @rproc: the rproc resource to release
+ *
+ * Releases the PRU rproc resource and makes it available to other
+ * users.
+ */
+void pru_rproc_put(struct rproc *rproc)
+{
+	struct pru_rproc *pru;
+
+	if (IS_ERR_OR_NULL(rproc) || !is_pru_rproc(rproc->dev.parent))
+		return;
+
+	pru = rproc->priv;
+
+	mutex_lock(&pru->lock);
+
+	if (!pru->client_np) {
+		mutex_unlock(&pru->lock);
+		return;
+	}
+
+	pru->client_np = NULL;
+	mutex_unlock(&pru->lock);
+
+	put_device(&rproc->dev);
+}
+EXPORT_SYMBOL_GPL(pru_rproc_put);
+
 static inline u32 pru_debug_read_reg(struct pru_rproc *pru, unsigned int reg)
 {
 	return readl_relaxed(pru->mem_regions[PRU_IOMEM_DEBUG].va + reg);
@@ -438,7 +563,7 @@ static void *pru_d_da_to_va(struct pru_rproc *pru, u32 da, size_t len)
 	dram0 = pruss->mem_regions[PRUSS_MEM_DRAM0];
 	dram1 = pruss->mem_regions[PRUSS_MEM_DRAM1];
 	/* PRU1 has its local RAM addresses reversed */
-	if (pru->id == 1)
+	if (pru->id == PRUSS_PRU1)
 		swap(dram0, dram1);
 	shrd_ram = pruss->mem_regions[PRUSS_MEM_SHRD_RAM2];
 
@@ -747,14 +872,14 @@ static int pru_rproc_set_id(struct pru_rproc *pru)
 	case RTU0_IRAM_ADDR_MASK:
 		fallthrough;
 	case PRU0_IRAM_ADDR_MASK:
-		pru->id = 0;
+		pru->id = PRUSS_PRU0;
 		break;
 	case TX_PRU1_IRAM_ADDR_MASK:
 		fallthrough;
 	case RTU1_IRAM_ADDR_MASK:
 		fallthrough;
 	case PRU1_IRAM_ADDR_MASK:
-		pru->id = 1;
+		pru->id = PRUSS_PRU1;
 		break;
 	default:
 		ret = -EINVAL;
@@ -816,6 +941,8 @@ static int pru_rproc_probe(struct platform_device *pdev)
 	pru->pruss = platform_get_drvdata(ppdev);
 	pru->rproc = rproc;
 	pru->fw_name = fw_name;
+	pru->client_np = NULL;
+	mutex_init(&pru->lock);
 
 	for (i = 0; i < ARRAY_SIZE(mem_names); i++) {
 		res = platform_get_resource_byname(pdev, IORESOURCE_MEM,
@@ -903,7 +1030,7 @@ MODULE_DEVICE_TABLE(of, pru_rproc_match);
 
 static struct platform_driver pru_rproc_driver = {
 	.driver = {
-		.name   = "pru-rproc",
+		.name   = PRU_RPROC_DRVNAME,
 		.of_match_table = pru_rproc_match,
 		.suppress_bind_attrs = true,
 	},
@@ -915,5 +1042,6 @@ module_platform_driver(pru_rproc_driver);
 MODULE_AUTHOR("Suman Anna <s-anna@ti.com>");
 MODULE_AUTHOR("Andrew F. Davis <afd@ti.com>");
 MODULE_AUTHOR("Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>");
+MODULE_AUTHOR("Puranjay Mohan <p-mohan@ti.com>");
 MODULE_DESCRIPTION("PRU-ICSS Remote Processor Driver");
 MODULE_LICENSE("GPL v2");
diff --git a/include/linux/pruss.h b/include/linux/pruss.h
new file mode 100644
index 000000000000..fdc719b43db0
--- /dev/null
+++ b/include/linux/pruss.h
@@ -0,0 +1,56 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/**
+ * PRU-ICSS Subsystem user interfaces
+ *
+ * Copyright (C) 2015-2022 Texas Instruments Incorporated - http://www.ti.com
+ *	Suman Anna <s-anna@ti.com>
+ */
+
+#ifndef __LINUX_PRUSS_H
+#define __LINUX_PRUSS_H
+
+#include <linux/device.h>
+#include <linux/types.h>
+
+#define PRU_RPROC_DRVNAME "pru-rproc"
+
+/*
+ * enum pruss_pru_id - PRU core identifiers
+ */
+enum pruss_pru_id {
+	PRUSS_PRU0 = 0,
+	PRUSS_PRU1,
+	PRUSS_NUM_PRUS,
+};
+
+struct device_node;
+
+#if IS_ENABLED(CONFIG_PRU_REMOTEPROC)
+
+struct rproc *pru_rproc_get(struct device_node *np, int index,
+			    enum pruss_pru_id *pru_id);
+void pru_rproc_put(struct rproc *rproc);
+
+#else
+
+static inline struct rproc *
+pru_rproc_get(struct device_node *np, int index, enum pruss_pru_id *pru_id)
+{
+	return ERR_PTR(-EOPNOTSUPP);
+}
+
+static inline void pru_rproc_put(struct rproc *rproc) { }
+
+#endif /* CONFIG_PRU_REMOTEPROC */
+
+static inline bool is_pru_rproc(struct device *dev)
+{
+	const char *drv_name = dev_driver_string(dev);
+
+	if (strncmp(drv_name, PRU_RPROC_DRVNAME, sizeof(PRU_RPROC_DRVNAME)))
+		return false;
+
+	return true;
+}
+
+#endif /* __LINUX_PRUSS_H */
-- 
2.17.1

Re: [PATCH v5 2/6] remoteproc: pru: Add APIs to get and put the PRU cores

[Mathieu Poirier]

Hi Puranjay,

On Tue, Jun 07, 2022 at 10:26:46AM +0530, Puranjay Mohan wrote:
> From: Tero Kristo <t-kristo@ti.com>
> 
> Add two new APIs, pru_rproc_get() and pru_rproc_put(), to the PRU
> driver to allow client drivers to acquire and release the remoteproc
> device associated with a PRU core. The PRU cores are treated as
> resources with only one client owning it at a time.
> 
> The pru_rproc_get() function returns the rproc handle corresponding
> to a PRU core identified by the device tree "ti,prus" property under
> the client node. The pru_rproc_put() is the complementary function
> to pru_rproc_get().
> 
> Co-developed-by: Suman Anna <s-anna@ti.com>
> Signed-off-by: Suman Anna <s-anna@ti.com>
> Signed-off-by: Tero Kristo <t-kristo@ti.com>
> Co-developed-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
> Signed-off-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
> Co-developed-by: Puranjay Mohan <p-mohan@ti.com>
> Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
> ---
>  drivers/remoteproc/pru_rproc.c | 138 +++++++++++++++++++++++++++++++--
>  include/linux/pruss.h          |  56 +++++++++++++
>  2 files changed, 189 insertions(+), 5 deletions(-)
>  create mode 100644 include/linux/pruss.h
> 
> diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
> index 1777a01fa84e..7a35b400287a 100644
> --- a/drivers/remoteproc/pru_rproc.c
> +++ b/drivers/remoteproc/pru_rproc.c
> @@ -2,12 +2,13 @@
>  /*
>   * PRU-ICSS remoteproc driver for various TI SoCs
>   *
> - * Copyright (C) 2014-2020 Texas Instruments Incorporated - https://www.ti.com/
> + * Copyright (C) 2014-2022 Texas Instruments Incorporated - https://www.ti.com/
>   *
>   * Author(s):
>   *	Suman Anna <s-anna@ti.com>
>   *	Andrew F. Davis <afd@ti.com>
>   *	Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org> for Texas Instruments
> + *	Puranjay Mohan <p-mohan@ti.com>
>   */
>  
>  #include <linux/bitops.h>
> @@ -16,6 +17,7 @@
>  #include <linux/module.h>
>  #include <linux/of_device.h>
>  #include <linux/of_irq.h>
> +#include <linux/pruss.h>
>  #include <linux/pruss_driver.h>
>  #include <linux/remoteproc.h>
>  
> @@ -111,6 +113,8 @@ struct pru_private_data {
>   * @rproc: remoteproc pointer for this PRU core
>   * @data: PRU core specific data
>   * @mem_regions: data for each of the PRU memory regions
> + * @client_np: client device node
> + * @lock: mutex to protect client usage
>   * @fw_name: name of firmware image used during loading
>   * @mapped_irq: virtual interrupt numbers of created fw specific mapping
>   * @pru_interrupt_map: pointer to interrupt mapping description (firmware)
> @@ -126,6 +130,8 @@ struct pru_rproc {
>  	struct rproc *rproc;
>  	const struct pru_private_data *data;
>  	struct pruss_mem_region mem_regions[PRU_IOMEM_MAX];
> +	struct device_node *client_np;
> +	struct mutex lock; /* client access lock */
>  	const char *fw_name;
>  	unsigned int *mapped_irq;
>  	struct pru_irq_rsc *pru_interrupt_map;
> @@ -146,6 +152,125 @@ void pru_control_write_reg(struct pru_rproc *pru, unsigned int reg, u32 val)
>  	writel_relaxed(val, pru->mem_regions[PRU_IOMEM_CTRL].va + reg);
>  }
>  
> +static struct rproc *__pru_rproc_get(struct device_node *np, int index)
> +{
> +	struct device_node *rproc_np = NULL;
> +	struct platform_device *pdev;
> +	struct rproc *rproc;
> +
> +	rproc_np = of_parse_phandle(np, "ti,prus", index);
> +	if (!rproc_np || !of_device_is_available(rproc_np))
> +		return ERR_PTR(-ENODEV);
> +
> +	pdev = of_find_device_by_node(rproc_np);
> +	of_node_put(rproc_np);
> +
> +	if (!pdev || !(&pdev->dev) || !((&pdev->dev)->driver))
> +		/* probably PRU not yet probed */
> +		return ERR_PTR(-EPROBE_DEFER);
> +
> +	/* make sure it is PRU rproc */
> +	if (!is_pru_rproc(&pdev->dev)) {
> +		put_device(&pdev->dev);
> +		return ERR_PTR(-ENODEV);
> +	}
> +
> +	rproc = platform_get_drvdata(pdev);
> +	put_device(&pdev->dev);
> +	if (!rproc)
> +		return ERR_PTR(-EPROBE_DEFER);
> +
> +	get_device(&rproc->dev);
> +
> +	return rproc;
> +}
> +
> +/**
> + * pru_rproc_get() - get the PRU rproc instance from a device node
> + * @np: the user/client device node
> + * @index: index to use for the ti,prus property
> + * @pru_id: optional pointer to return the PRU remoteproc processor id
> + *
> + * This function looks through a client device node's "ti,prus" property at
> + * index @index and returns the rproc handle for a valid PRU remote processor if
> + * found. The function allows only one user to own the PRU rproc resource at a
> + * time. Caller must call pru_rproc_put() when done with using the rproc, not
> + * required if the function returns a failure.
> + *
> + * When optional @pru_id pointer is passed the PRU remoteproc processor id is
> + * returned.
> + *
> + * Return: rproc handle on success, and an ERR_PTR on failure using one
> + * of the following error values
> + *    -ENODEV if device is not found
> + *    -EBUSY if PRU is already acquired by anyone
> + *    -EPROBE_DEFER is PRU device is not probed yet
> + */
> +struct rproc *pru_rproc_get(struct device_node *np, int index,
> +			    enum pruss_pru_id *pru_id)
> +{
> +	struct rproc *rproc;
> +	struct pru_rproc *pru;
> +	struct device *dev;
> +
> +	try_module_get(THIS_MODULE);

There should be a module_put() in pru_rproc_put()...

More comments to come tomorrow.  I'm especially worried about this API racing
with a remote processor being removed or detached.

Thanks,
Mathieu

> +
> +	rproc = __pru_rproc_get(np, index);
> +	if (IS_ERR(rproc))
> +		return rproc;
> +
> +	pru = rproc->priv;
> +	dev = &rproc->dev;
> +
> +	mutex_lock(&pru->lock);
> +
> +	if (pru->client_np) {
> +		mutex_unlock(&pru->lock);
> +		put_device(dev);
> +		return ERR_PTR(-EBUSY);
> +	}
> +
> +	pru->client_np = np;
> +
> +	mutex_unlock(&pru->lock);
> +
> +	if (pru_id)
> +		*pru_id = pru->id;
> +
> +	return rproc;
> +}
> +EXPORT_SYMBOL_GPL(pru_rproc_get);
> +
> +/**
> + * pru_rproc_put() - release the PRU rproc resource
> + * @rproc: the rproc resource to release
> + *
> + * Releases the PRU rproc resource and makes it available to other
> + * users.
> + */
> +void pru_rproc_put(struct rproc *rproc)
> +{
> +	struct pru_rproc *pru;
> +
> +	if (IS_ERR_OR_NULL(rproc) || !is_pru_rproc(rproc->dev.parent))
> +		return;
> +
> +	pru = rproc->priv;
> +
> +	mutex_lock(&pru->lock);
> +
> +	if (!pru->client_np) {
> +		mutex_unlock(&pru->lock);
> +		return;
> +	}
> +
> +	pru->client_np = NULL;
> +	mutex_unlock(&pru->lock);
> +
> +	put_device(&rproc->dev);
> +}
> +EXPORT_SYMBOL_GPL(pru_rproc_put);
> +
>  static inline u32 pru_debug_read_reg(struct pru_rproc *pru, unsigned int reg)
>  {
>  	return readl_relaxed(pru->mem_regions[PRU_IOMEM_DEBUG].va + reg);
> @@ -438,7 +563,7 @@ static void *pru_d_da_to_va(struct pru_rproc *pru, u32 da, size_t len)
>  	dram0 = pruss->mem_regions[PRUSS_MEM_DRAM0];
>  	dram1 = pruss->mem_regions[PRUSS_MEM_DRAM1];
>  	/* PRU1 has its local RAM addresses reversed */
> -	if (pru->id == 1)
> +	if (pru->id == PRUSS_PRU1)
>  		swap(dram0, dram1);
>  	shrd_ram = pruss->mem_regions[PRUSS_MEM_SHRD_RAM2];
>  
> @@ -747,14 +872,14 @@ static int pru_rproc_set_id(struct pru_rproc *pru)
>  	case RTU0_IRAM_ADDR_MASK:
>  		fallthrough;
>  	case PRU0_IRAM_ADDR_MASK:
> -		pru->id = 0;
> +		pru->id = PRUSS_PRU0;
>  		break;
>  	case TX_PRU1_IRAM_ADDR_MASK:
>  		fallthrough;
>  	case RTU1_IRAM_ADDR_MASK:
>  		fallthrough;
>  	case PRU1_IRAM_ADDR_MASK:
> -		pru->id = 1;
> +		pru->id = PRUSS_PRU1;
>  		break;
>  	default:
>  		ret = -EINVAL;
> @@ -816,6 +941,8 @@ static int pru_rproc_probe(struct platform_device *pdev)
>  	pru->pruss = platform_get_drvdata(ppdev);
>  	pru->rproc = rproc;
>  	pru->fw_name = fw_name;
> +	pru->client_np = NULL;
> +	mutex_init(&pru->lock);
>  
>  	for (i = 0; i < ARRAY_SIZE(mem_names); i++) {
>  		res = platform_get_resource_byname(pdev, IORESOURCE_MEM,
> @@ -903,7 +1030,7 @@ MODULE_DEVICE_TABLE(of, pru_rproc_match);
>  
>  static struct platform_driver pru_rproc_driver = {
>  	.driver = {
> -		.name   = "pru-rproc",
> +		.name   = PRU_RPROC_DRVNAME,
>  		.of_match_table = pru_rproc_match,
>  		.suppress_bind_attrs = true,
>  	},
> @@ -915,5 +1042,6 @@ module_platform_driver(pru_rproc_driver);
>  MODULE_AUTHOR("Suman Anna <s-anna@ti.com>");
>  MODULE_AUTHOR("Andrew F. Davis <afd@ti.com>");
>  MODULE_AUTHOR("Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>");
> +MODULE_AUTHOR("Puranjay Mohan <p-mohan@ti.com>");
>  MODULE_DESCRIPTION("PRU-ICSS Remote Processor Driver");
>  MODULE_LICENSE("GPL v2");
> diff --git a/include/linux/pruss.h b/include/linux/pruss.h
> new file mode 100644
> index 000000000000..fdc719b43db0
> --- /dev/null
> +++ b/include/linux/pruss.h
> @@ -0,0 +1,56 @@
> +/* SPDX-License-Identifier: GPL-2.0-only */
> +/**
> + * PRU-ICSS Subsystem user interfaces
> + *
> + * Copyright (C) 2015-2022 Texas Instruments Incorporated - http://www.ti.com
> + *	Suman Anna <s-anna@ti.com>
> + */
> +
> +#ifndef __LINUX_PRUSS_H
> +#define __LINUX_PRUSS_H
> +
> +#include <linux/device.h>
> +#include <linux/types.h>
> +
> +#define PRU_RPROC_DRVNAME "pru-rproc"
> +
> +/*
> + * enum pruss_pru_id - PRU core identifiers
> + */
> +enum pruss_pru_id {
> +	PRUSS_PRU0 = 0,
> +	PRUSS_PRU1,
> +	PRUSS_NUM_PRUS,
> +};
> +
> +struct device_node;
> +
> +#if IS_ENABLED(CONFIG_PRU_REMOTEPROC)
> +
> +struct rproc *pru_rproc_get(struct device_node *np, int index,
> +			    enum pruss_pru_id *pru_id);
> +void pru_rproc_put(struct rproc *rproc);
> +
> +#else
> +
> +static inline struct rproc *
> +pru_rproc_get(struct device_node *np, int index, enum pruss_pru_id *pru_id)
> +{
> +	return ERR_PTR(-EOPNOTSUPP);
> +}
> +
> +static inline void pru_rproc_put(struct rproc *rproc) { }
> +
> +#endif /* CONFIG_PRU_REMOTEPROC */
> +
> +static inline bool is_pru_rproc(struct device *dev)
> +{
> +	const char *drv_name = dev_driver_string(dev);
> +
> +	if (strncmp(drv_name, PRU_RPROC_DRVNAME, sizeof(PRU_RPROC_DRVNAME)))
> +		return false;
> +
> +	return true;
> +}
> +
> +#endif /* __LINUX_PRUSS_H */
> -- 
> 2.17.1
> 

Re: [PATCH v5 2/6] remoteproc: pru: Add APIs to get and put the PRU cores

[Mathieu Poirier]

On Tue, Jul 12, 2022 at 11:59:49AM -0600, Mathieu Poirier wrote:
> Hi Puranjay,
> 
> On Tue, Jun 07, 2022 at 10:26:46AM +0530, Puranjay Mohan wrote:
> > From: Tero Kristo <t-kristo@ti.com>
> > 
> > Add two new APIs, pru_rproc_get() and pru_rproc_put(), to the PRU
> > driver to allow client drivers to acquire and release the remoteproc
> > device associated with a PRU core. The PRU cores are treated as
> > resources with only one client owning it at a time.
> > 
> > The pru_rproc_get() function returns the rproc handle corresponding
> > to a PRU core identified by the device tree "ti,prus" property under
> > the client node. The pru_rproc_put() is the complementary function
> > to pru_rproc_get().
> > 
> > Co-developed-by: Suman Anna <s-anna@ti.com>
> > Signed-off-by: Suman Anna <s-anna@ti.com>
> > Signed-off-by: Tero Kristo <t-kristo@ti.com>
> > Co-developed-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
> > Signed-off-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
> > Co-developed-by: Puranjay Mohan <p-mohan@ti.com>
> > Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
> > ---
> >  drivers/remoteproc/pru_rproc.c | 138 +++++++++++++++++++++++++++++++--
> >  include/linux/pruss.h          |  56 +++++++++++++
> >  2 files changed, 189 insertions(+), 5 deletions(-)
> >  create mode 100644 include/linux/pruss.h
> > 
> > diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
> > index 1777a01fa84e..7a35b400287a 100644
> > --- a/drivers/remoteproc/pru_rproc.c
> > +++ b/drivers/remoteproc/pru_rproc.c
> > @@ -2,12 +2,13 @@
> >  /*
> >   * PRU-ICSS remoteproc driver for various TI SoCs
> >   *
> > - * Copyright (C) 2014-2020 Texas Instruments Incorporated - https://www.ti.com/
> > + * Copyright (C) 2014-2022 Texas Instruments Incorporated - https://www.ti.com/
> >   *
> >   * Author(s):
> >   *	Suman Anna <s-anna@ti.com>
> >   *	Andrew F. Davis <afd@ti.com>
> >   *	Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org> for Texas Instruments
> > + *	Puranjay Mohan <p-mohan@ti.com>
> >   */
> >  
> >  #include <linux/bitops.h>
> > @@ -16,6 +17,7 @@
> >  #include <linux/module.h>
> >  #include <linux/of_device.h>
> >  #include <linux/of_irq.h>
> > +#include <linux/pruss.h>
> >  #include <linux/pruss_driver.h>
> >  #include <linux/remoteproc.h>
> >  
> > @@ -111,6 +113,8 @@ struct pru_private_data {
> >   * @rproc: remoteproc pointer for this PRU core
> >   * @data: PRU core specific data
> >   * @mem_regions: data for each of the PRU memory regions
> > + * @client_np: client device node
> > + * @lock: mutex to protect client usage
> >   * @fw_name: name of firmware image used during loading
> >   * @mapped_irq: virtual interrupt numbers of created fw specific mapping
> >   * @pru_interrupt_map: pointer to interrupt mapping description (firmware)
> > @@ -126,6 +130,8 @@ struct pru_rproc {
> >  	struct rproc *rproc;
> >  	const struct pru_private_data *data;
> >  	struct pruss_mem_region mem_regions[PRU_IOMEM_MAX];
> > +	struct device_node *client_np;
> > +	struct mutex lock; /* client access lock */
> >  	const char *fw_name;
> >  	unsigned int *mapped_irq;
> >  	struct pru_irq_rsc *pru_interrupt_map;
> > @@ -146,6 +152,125 @@ void pru_control_write_reg(struct pru_rproc *pru, unsigned int reg, u32 val)
> >  	writel_relaxed(val, pru->mem_regions[PRU_IOMEM_CTRL].va + reg);
> >  }
> >  
> > +static struct rproc *__pru_rproc_get(struct device_node *np, int index)
> > +{
> > +	struct device_node *rproc_np = NULL;
> > +	struct platform_device *pdev;
> > +	struct rproc *rproc;
> > +
> > +	rproc_np = of_parse_phandle(np, "ti,prus", index);
> > +	if (!rproc_np || !of_device_is_available(rproc_np))
> > +		return ERR_PTR(-ENODEV);
> > +
> > +	pdev = of_find_device_by_node(rproc_np);
> > +	of_node_put(rproc_np);
> > +
> > +	if (!pdev || !(&pdev->dev) || !((&pdev->dev)->driver))
> > +		/* probably PRU not yet probed */
> > +		return ERR_PTR(-EPROBE_DEFER);
> > +
> > +	/* make sure it is PRU rproc */
> > +	if (!is_pru_rproc(&pdev->dev)) {
> > +		put_device(&pdev->dev);
> > +		return ERR_PTR(-ENODEV);
> > +	}
> > +
> > +	rproc = platform_get_drvdata(pdev);
> > +	put_device(&pdev->dev);
> > +	if (!rproc)
> > +		return ERR_PTR(-EPROBE_DEFER);
> > +
> > +	get_device(&rproc->dev);
> > +
> > +	return rproc;
> > +}
> > +
> > +/**
> > + * pru_rproc_get() - get the PRU rproc instance from a device node
> > + * @np: the user/client device node
> > + * @index: index to use for the ti,prus property
> > + * @pru_id: optional pointer to return the PRU remoteproc processor id
> > + *
> > + * This function looks through a client device node's "ti,prus" property at
> > + * index @index and returns the rproc handle for a valid PRU remote processor if
> > + * found. The function allows only one user to own the PRU rproc resource at a
> > + * time. Caller must call pru_rproc_put() when done with using the rproc, not
> > + * required if the function returns a failure.
> > + *
> > + * When optional @pru_id pointer is passed the PRU remoteproc processor id is
> > + * returned.
> > + *
> > + * Return: rproc handle on success, and an ERR_PTR on failure using one
> > + * of the following error values
> > + *    -ENODEV if device is not found
> > + *    -EBUSY if PRU is already acquired by anyone
> > + *    -EPROBE_DEFER is PRU device is not probed yet
> > + */
> > +struct rproc *pru_rproc_get(struct device_node *np, int index,
> > +			    enum pruss_pru_id *pru_id)
> > +{
> > +	struct rproc *rproc;
> > +	struct pru_rproc *pru;
> > +	struct device *dev;
> > +
> > +	try_module_get(THIS_MODULE);
> 
> There should be a module_put() in pru_rproc_put()...

... and in the error path of this function.

> 
> More comments to come tomorrow.  I'm especially worried about this API racing
> with a remote processor being removed or detached.
>

Looking at what is done in wkup_m3_ipc_probe(), it should be possible to call
rproc_get_by_handle() here and that would make sure the remote processor doesn't
go away before the end of the function.

More comments to come...


> Thanks,
> Mathieu
> 
> > +
> > +	rproc = __pru_rproc_get(np, index);
> > +	if (IS_ERR(rproc))
> > +		return rproc;
> > +
> > +	pru = rproc->priv;
> > +	dev = &rproc->dev;
> > +
> > +	mutex_lock(&pru->lock);
> > +
> > +	if (pru->client_np) {
> > +		mutex_unlock(&pru->lock);
> > +		put_device(dev);
> > +		return ERR_PTR(-EBUSY);
> > +	}
> > +
> > +	pru->client_np = np;
> > +
> > +	mutex_unlock(&pru->lock);
> > +
> > +	if (pru_id)
> > +		*pru_id = pru->id;
> > +
> > +	return rproc;
> > +}
> > +EXPORT_SYMBOL_GPL(pru_rproc_get);
> > +
> > +/**
> > + * pru_rproc_put() - release the PRU rproc resource
> > + * @rproc: the rproc resource to release
> > + *
> > + * Releases the PRU rproc resource and makes it available to other
> > + * users.
> > + */
> > +void pru_rproc_put(struct rproc *rproc)
> > +{
> > +	struct pru_rproc *pru;
> > +
> > +	if (IS_ERR_OR_NULL(rproc) || !is_pru_rproc(rproc->dev.parent))
> > +		return;
> > +
> > +	pru = rproc->priv;
> > +
> > +	mutex_lock(&pru->lock);
> > +
> > +	if (!pru->client_np) {
> > +		mutex_unlock(&pru->lock);
> > +		return;
> > +	}
> > +
> > +	pru->client_np = NULL;
> > +	mutex_unlock(&pru->lock);
> > +
> > +	put_device(&rproc->dev);
> > +}
> > +EXPORT_SYMBOL_GPL(pru_rproc_put);
> > +
> >  static inline u32 pru_debug_read_reg(struct pru_rproc *pru, unsigned int reg)
> >  {
> >  	return readl_relaxed(pru->mem_regions[PRU_IOMEM_DEBUG].va + reg);
> > @@ -438,7 +563,7 @@ static void *pru_d_da_to_va(struct pru_rproc *pru, u32 da, size_t len)
> >  	dram0 = pruss->mem_regions[PRUSS_MEM_DRAM0];
> >  	dram1 = pruss->mem_regions[PRUSS_MEM_DRAM1];
> >  	/* PRU1 has its local RAM addresses reversed */
> > -	if (pru->id == 1)
> > +	if (pru->id == PRUSS_PRU1)
> >  		swap(dram0, dram1);
> >  	shrd_ram = pruss->mem_regions[PRUSS_MEM_SHRD_RAM2];
> >  
> > @@ -747,14 +872,14 @@ static int pru_rproc_set_id(struct pru_rproc *pru)
> >  	case RTU0_IRAM_ADDR_MASK:
> >  		fallthrough;
> >  	case PRU0_IRAM_ADDR_MASK:
> > -		pru->id = 0;
> > +		pru->id = PRUSS_PRU0;
> >  		break;
> >  	case TX_PRU1_IRAM_ADDR_MASK:
> >  		fallthrough;
> >  	case RTU1_IRAM_ADDR_MASK:
> >  		fallthrough;
> >  	case PRU1_IRAM_ADDR_MASK:
> > -		pru->id = 1;
> > +		pru->id = PRUSS_PRU1;
> >  		break;
> >  	default:
> >  		ret = -EINVAL;
> > @@ -816,6 +941,8 @@ static int pru_rproc_probe(struct platform_device *pdev)
> >  	pru->pruss = platform_get_drvdata(ppdev);
> >  	pru->rproc = rproc;
> >  	pru->fw_name = fw_name;
> > +	pru->client_np = NULL;
> > +	mutex_init(&pru->lock);
> >  
> >  	for (i = 0; i < ARRAY_SIZE(mem_names); i++) {
> >  		res = platform_get_resource_byname(pdev, IORESOURCE_MEM,
> > @@ -903,7 +1030,7 @@ MODULE_DEVICE_TABLE(of, pru_rproc_match);
> >  
> >  static struct platform_driver pru_rproc_driver = {
> >  	.driver = {
> > -		.name   = "pru-rproc",
> > +		.name   = PRU_RPROC_DRVNAME,
> >  		.of_match_table = pru_rproc_match,
> >  		.suppress_bind_attrs = true,
> >  	},
> > @@ -915,5 +1042,6 @@ module_platform_driver(pru_rproc_driver);
> >  MODULE_AUTHOR("Suman Anna <s-anna@ti.com>");
> >  MODULE_AUTHOR("Andrew F. Davis <afd@ti.com>");
> >  MODULE_AUTHOR("Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>");
> > +MODULE_AUTHOR("Puranjay Mohan <p-mohan@ti.com>");
> >  MODULE_DESCRIPTION("PRU-ICSS Remote Processor Driver");
> >  MODULE_LICENSE("GPL v2");
> > diff --git a/include/linux/pruss.h b/include/linux/pruss.h
> > new file mode 100644
> > index 000000000000..fdc719b43db0
> > --- /dev/null
> > +++ b/include/linux/pruss.h
> > @@ -0,0 +1,56 @@
> > +/* SPDX-License-Identifier: GPL-2.0-only */
> > +/**
> > + * PRU-ICSS Subsystem user interfaces
> > + *
> > + * Copyright (C) 2015-2022 Texas Instruments Incorporated - http://www.ti.com
> > + *	Suman Anna <s-anna@ti.com>
> > + */
> > +
> > +#ifndef __LINUX_PRUSS_H
> > +#define __LINUX_PRUSS_H
> > +
> > +#include <linux/device.h>
> > +#include <linux/types.h>
> > +
> > +#define PRU_RPROC_DRVNAME "pru-rproc"
> > +
> > +/*
> > + * enum pruss_pru_id - PRU core identifiers
> > + */
> > +enum pruss_pru_id {
> > +	PRUSS_PRU0 = 0,
> > +	PRUSS_PRU1,
> > +	PRUSS_NUM_PRUS,
> > +};
> > +
> > +struct device_node;
> > +
> > +#if IS_ENABLED(CONFIG_PRU_REMOTEPROC)
> > +
> > +struct rproc *pru_rproc_get(struct device_node *np, int index,
> > +			    enum pruss_pru_id *pru_id);
> > +void pru_rproc_put(struct rproc *rproc);
> > +
> > +#else
> > +
> > +static inline struct rproc *
> > +pru_rproc_get(struct device_node *np, int index, enum pruss_pru_id *pru_id)
> > +{
> > +	return ERR_PTR(-EOPNOTSUPP);
> > +}
> > +
> > +static inline void pru_rproc_put(struct rproc *rproc) { }
> > +
> > +#endif /* CONFIG_PRU_REMOTEPROC */
> > +
> > +static inline bool is_pru_rproc(struct device *dev)
> > +{
> > +	const char *drv_name = dev_driver_string(dev);
> > +
> > +	if (strncmp(drv_name, PRU_RPROC_DRVNAME, sizeof(PRU_RPROC_DRVNAME)))
> > +		return false;
> > +
> > +	return true;
> > +}
> > +
> > +#endif /* __LINUX_PRUSS_H */
> > -- 
> > 2.17.1
> > 

Re: [PATCH v5 2/6] remoteproc: pru: Add APIs to get and put the PRU cores

[Kishon Vijay Abraham I]

+Danish

Hi Mathieu,

On 14/07/22 22:50, Mathieu Poirier wrote:
> On Tue, Jul 12, 2022 at 11:59:49AM -0600, Mathieu Poirier wrote:
>> Hi Puranjay,

Removed Puranjay (as he is no longer with TI) and adding Danish.

Regards,
Kishon
>>
>> On Tue, Jun 07, 2022 at 10:26:46AM +0530, Puranjay Mohan wrote:
>>> From: Tero Kristo <t-kristo@ti.com>
>>>
>>> Add two new APIs, pru_rproc_get() and pru_rproc_put(), to the PRU
>>> driver to allow client drivers to acquire and release the remoteproc
>>> device associated with a PRU core. The PRU cores are treated as
>>> resources with only one client owning it at a time.
>>>
>>> The pru_rproc_get() function returns the rproc handle corresponding
>>> to a PRU core identified by the device tree "ti,prus" property under
>>> the client node. The pru_rproc_put() is the complementary function
>>> to pru_rproc_get().
>>>
>>> Co-developed-by: Suman Anna <s-anna@ti.com>
>>> Signed-off-by: Suman Anna <s-anna@ti.com>
>>> Signed-off-by: Tero Kristo <t-kristo@ti.com>
>>> Co-developed-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
>>> Signed-off-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
>>> Co-developed-by: Puranjay Mohan <p-mohan@ti.com>
>>> Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
>>> ---
>>>  drivers/remoteproc/pru_rproc.c | 138 +++++++++++++++++++++++++++++++--
>>>  include/linux/pruss.h          |  56 +++++++++++++
>>>  2 files changed, 189 insertions(+), 5 deletions(-)
>>>  create mode 100644 include/linux/pruss.h
>>>
>>> diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
>>> index 1777a01fa84e..7a35b400287a 100644
>>> --- a/drivers/remoteproc/pru_rproc.c
>>> +++ b/drivers/remoteproc/pru_rproc.c
>>> @@ -2,12 +2,13 @@
>>>  /*
>>>   * PRU-ICSS remoteproc driver for various TI SoCs
>>>   *
>>> - * Copyright (C) 2014-2020 Texas Instruments Incorporated - https://www.ti.com/
>>> + * Copyright (C) 2014-2022 Texas Instruments Incorporated - https://www.ti.com/
>>>   *
>>>   * Author(s):
>>>   *	Suman Anna <s-anna@ti.com>
>>>   *	Andrew F. Davis <afd@ti.com>
>>>   *	Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org> for Texas Instruments
>>> + *	Puranjay Mohan <p-mohan@ti.com>
>>>   */
>>>  
>>>  #include <linux/bitops.h>
>>> @@ -16,6 +17,7 @@
>>>  #include <linux/module.h>
>>>  #include <linux/of_device.h>
>>>  #include <linux/of_irq.h>
>>> +#include <linux/pruss.h>
>>>  #include <linux/pruss_driver.h>
>>>  #include <linux/remoteproc.h>
>>>  
>>> @@ -111,6 +113,8 @@ struct pru_private_data {
>>>   * @rproc: remoteproc pointer for this PRU core
>>>   * @data: PRU core specific data
>>>   * @mem_regions: data for each of the PRU memory regions
>>> + * @client_np: client device node
>>> + * @lock: mutex to protect client usage
>>>   * @fw_name: name of firmware image used during loading
>>>   * @mapped_irq: virtual interrupt numbers of created fw specific mapping
>>>   * @pru_interrupt_map: pointer to interrupt mapping description (firmware)
>>> @@ -126,6 +130,8 @@ struct pru_rproc {
>>>  	struct rproc *rproc;
>>>  	const struct pru_private_data *data;
>>>  	struct pruss_mem_region mem_regions[PRU_IOMEM_MAX];
>>> +	struct device_node *client_np;
>>> +	struct mutex lock; /* client access lock */
>>>  	const char *fw_name;
>>>  	unsigned int *mapped_irq;
>>>  	struct pru_irq_rsc *pru_interrupt_map;
>>> @@ -146,6 +152,125 @@ void pru_control_write_reg(struct pru_rproc *pru, unsigned int reg, u32 val)
>>>  	writel_relaxed(val, pru->mem_regions[PRU_IOMEM_CTRL].va + reg);
>>>  }
>>>  
>>> +static struct rproc *__pru_rproc_get(struct device_node *np, int index)
>>> +{
>>> +	struct device_node *rproc_np = NULL;
>>> +	struct platform_device *pdev;
>>> +	struct rproc *rproc;
>>> +
>>> +	rproc_np = of_parse_phandle(np, "ti,prus", index);
>>> +	if (!rproc_np || !of_device_is_available(rproc_np))
>>> +		return ERR_PTR(-ENODEV);
>>> +
>>> +	pdev = of_find_device_by_node(rproc_np);
>>> +	of_node_put(rproc_np);
>>> +
>>> +	if (!pdev || !(&pdev->dev) || !((&pdev->dev)->driver))
>>> +		/* probably PRU not yet probed */
>>> +		return ERR_PTR(-EPROBE_DEFER);
>>> +
>>> +	/* make sure it is PRU rproc */
>>> +	if (!is_pru_rproc(&pdev->dev)) {
>>> +		put_device(&pdev->dev);
>>> +		return ERR_PTR(-ENODEV);
>>> +	}
>>> +
>>> +	rproc = platform_get_drvdata(pdev);
>>> +	put_device(&pdev->dev);
>>> +	if (!rproc)
>>> +		return ERR_PTR(-EPROBE_DEFER);
>>> +
>>> +	get_device(&rproc->dev);
>>> +
>>> +	return rproc;
>>> +}
>>> +
>>> +/**
>>> + * pru_rproc_get() - get the PRU rproc instance from a device node
>>> + * @np: the user/client device node
>>> + * @index: index to use for the ti,prus property
>>> + * @pru_id: optional pointer to return the PRU remoteproc processor id
>>> + *
>>> + * This function looks through a client device node's "ti,prus" property at
>>> + * index @index and returns the rproc handle for a valid PRU remote processor if
>>> + * found. The function allows only one user to own the PRU rproc resource at a
>>> + * time. Caller must call pru_rproc_put() when done with using the rproc, not
>>> + * required if the function returns a failure.
>>> + *
>>> + * When optional @pru_id pointer is passed the PRU remoteproc processor id is
>>> + * returned.
>>> + *
>>> + * Return: rproc handle on success, and an ERR_PTR on failure using one
>>> + * of the following error values
>>> + *    -ENODEV if device is not found
>>> + *    -EBUSY if PRU is already acquired by anyone
>>> + *    -EPROBE_DEFER is PRU device is not probed yet
>>> + */
>>> +struct rproc *pru_rproc_get(struct device_node *np, int index,
>>> +			    enum pruss_pru_id *pru_id)
>>> +{
>>> +	struct rproc *rproc;
>>> +	struct pru_rproc *pru;
>>> +	struct device *dev;
>>> +
>>> +	try_module_get(THIS_MODULE);
>>
>> There should be a module_put() in pru_rproc_put()...
> 
> ... and in the error path of this function.
> 
>>
>> More comments to come tomorrow.  I'm especially worried about this API racing
>> with a remote processor being removed or detached.
>>
> 
> Looking at what is done in wkup_m3_ipc_probe(), it should be possible to call
> rproc_get_by_handle() here and that would make sure the remote processor doesn't
> go away before the end of the function.
> 
> More comments to come...
> 
> 
>> Thanks,
>> Mathieu
>>
>>> +
>>> +	rproc = __pru_rproc_get(np, index);
>>> +	if (IS_ERR(rproc))
>>> +		return rproc;
>>> +
>>> +	pru = rproc->priv;
>>> +	dev = &rproc->dev;
>>> +
>>> +	mutex_lock(&pru->lock);
>>> +
>>> +	if (pru->client_np) {
>>> +		mutex_unlock(&pru->lock);
>>> +		put_device(dev);
>>> +		return ERR_PTR(-EBUSY);
>>> +	}
>>> +
>>> +	pru->client_np = np;
>>> +
>>> +	mutex_unlock(&pru->lock);
>>> +
>>> +	if (pru_id)
>>> +		*pru_id = pru->id;
>>> +
>>> +	return rproc;
>>> +}
>>> +EXPORT_SYMBOL_GPL(pru_rproc_get);
>>> +
>>> +/**
>>> + * pru_rproc_put() - release the PRU rproc resource
>>> + * @rproc: the rproc resource to release
>>> + *
>>> + * Releases the PRU rproc resource and makes it available to other
>>> + * users.
>>> + */
>>> +void pru_rproc_put(struct rproc *rproc)
>>> +{
>>> +	struct pru_rproc *pru;
>>> +
>>> +	if (IS_ERR_OR_NULL(rproc) || !is_pru_rproc(rproc->dev.parent))
>>> +		return;
>>> +
>>> +	pru = rproc->priv;
>>> +
>>> +	mutex_lock(&pru->lock);
>>> +
>>> +	if (!pru->client_np) {
>>> +		mutex_unlock(&pru->lock);
>>> +		return;
>>> +	}
>>> +
>>> +	pru->client_np = NULL;
>>> +	mutex_unlock(&pru->lock);
>>> +
>>> +	put_device(&rproc->dev);
>>> +}
>>> +EXPORT_SYMBOL_GPL(pru_rproc_put);
>>> +
>>>  static inline u32 pru_debug_read_reg(struct pru_rproc *pru, unsigned int reg)
>>>  {
>>>  	return readl_relaxed(pru->mem_regions[PRU_IOMEM_DEBUG].va + reg);
>>> @@ -438,7 +563,7 @@ static void *pru_d_da_to_va(struct pru_rproc *pru, u32 da, size_t len)
>>>  	dram0 = pruss->mem_regions[PRUSS_MEM_DRAM0];
>>>  	dram1 = pruss->mem_regions[PRUSS_MEM_DRAM1];
>>>  	/* PRU1 has its local RAM addresses reversed */
>>> -	if (pru->id == 1)
>>> +	if (pru->id == PRUSS_PRU1)
>>>  		swap(dram0, dram1);
>>>  	shrd_ram = pruss->mem_regions[PRUSS_MEM_SHRD_RAM2];
>>>  
>>> @@ -747,14 +872,14 @@ static int pru_rproc_set_id(struct pru_rproc *pru)
>>>  	case RTU0_IRAM_ADDR_MASK:
>>>  		fallthrough;
>>>  	case PRU0_IRAM_ADDR_MASK:
>>> -		pru->id = 0;
>>> +		pru->id = PRUSS_PRU0;
>>>  		break;
>>>  	case TX_PRU1_IRAM_ADDR_MASK:
>>>  		fallthrough;
>>>  	case RTU1_IRAM_ADDR_MASK:
>>>  		fallthrough;
>>>  	case PRU1_IRAM_ADDR_MASK:
>>> -		pru->id = 1;
>>> +		pru->id = PRUSS_PRU1;
>>>  		break;
>>>  	default:
>>>  		ret = -EINVAL;
>>> @@ -816,6 +941,8 @@ static int pru_rproc_probe(struct platform_device *pdev)
>>>  	pru->pruss = platform_get_drvdata(ppdev);
>>>  	pru->rproc = rproc;
>>>  	pru->fw_name = fw_name;
>>> +	pru->client_np = NULL;
>>> +	mutex_init(&pru->lock);
>>>  
>>>  	for (i = 0; i < ARRAY_SIZE(mem_names); i++) {
>>>  		res = platform_get_resource_byname(pdev, IORESOURCE_MEM,
>>> @@ -903,7 +1030,7 @@ MODULE_DEVICE_TABLE(of, pru_rproc_match);
>>>  
>>>  static struct platform_driver pru_rproc_driver = {
>>>  	.driver = {
>>> -		.name   = "pru-rproc",
>>> +		.name   = PRU_RPROC_DRVNAME,
>>>  		.of_match_table = pru_rproc_match,
>>>  		.suppress_bind_attrs = true,
>>>  	},
>>> @@ -915,5 +1042,6 @@ module_platform_driver(pru_rproc_driver);
>>>  MODULE_AUTHOR("Suman Anna <s-anna@ti.com>");
>>>  MODULE_AUTHOR("Andrew F. Davis <afd@ti.com>");
>>>  MODULE_AUTHOR("Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>");
>>> +MODULE_AUTHOR("Puranjay Mohan <p-mohan@ti.com>");
>>>  MODULE_DESCRIPTION("PRU-ICSS Remote Processor Driver");
>>>  MODULE_LICENSE("GPL v2");
>>> diff --git a/include/linux/pruss.h b/include/linux/pruss.h
>>> new file mode 100644
>>> index 000000000000..fdc719b43db0
>>> --- /dev/null
>>> +++ b/include/linux/pruss.h
>>> @@ -0,0 +1,56 @@
>>> +/* SPDX-License-Identifier: GPL-2.0-only */
>>> +/**
>>> + * PRU-ICSS Subsystem user interfaces
>>> + *
>>> + * Copyright (C) 2015-2022 Texas Instruments Incorporated - http://www.ti.com
>>> + *	Suman Anna <s-anna@ti.com>
>>> + */
>>> +
>>> +#ifndef __LINUX_PRUSS_H
>>> +#define __LINUX_PRUSS_H
>>> +
>>> +#include <linux/device.h>
>>> +#include <linux/types.h>
>>> +
>>> +#define PRU_RPROC_DRVNAME "pru-rproc"
>>> +
>>> +/*
>>> + * enum pruss_pru_id - PRU core identifiers
>>> + */
>>> +enum pruss_pru_id {
>>> +	PRUSS_PRU0 = 0,
>>> +	PRUSS_PRU1,
>>> +	PRUSS_NUM_PRUS,
>>> +};
>>> +
>>> +struct device_node;
>>> +
>>> +#if IS_ENABLED(CONFIG_PRU_REMOTEPROC)
>>> +
>>> +struct rproc *pru_rproc_get(struct device_node *np, int index,
>>> +			    enum pruss_pru_id *pru_id);
>>> +void pru_rproc_put(struct rproc *rproc);
>>> +
>>> +#else
>>> +
>>> +static inline struct rproc *
>>> +pru_rproc_get(struct device_node *np, int index, enum pruss_pru_id *pru_id)
>>> +{
>>> +	return ERR_PTR(-EOPNOTSUPP);
>>> +}
>>> +
>>> +static inline void pru_rproc_put(struct rproc *rproc) { }
>>> +
>>> +#endif /* CONFIG_PRU_REMOTEPROC */
>>> +
>>> +static inline bool is_pru_rproc(struct device *dev)
>>> +{
>>> +	const char *drv_name = dev_driver_string(dev);
>>> +
>>> +	if (strncmp(drv_name, PRU_RPROC_DRVNAME, sizeof(PRU_RPROC_DRVNAME)))
>>> +		return false;
>>> +
>>> +	return true;
>>> +}
>>> +
>>> +#endif /* __LINUX_PRUSS_H */
>>> -- 
>>> 2.17.1
>>>

Re: [PATCH v5 2/6] remoteproc: pru: Add APIs to get and put the PRU cores

[Md Danish Anwar]

Hi Mathieu,


On 15/07/22 11:52, Kishon Vijay Abraham I wrote:
> +Danish
> 
> Hi Mathieu,
> 
> On 14/07/22 22:50, Mathieu Poirier wrote:
>> On Tue, Jul 12, 2022 at 11:59:49AM -0600, Mathieu Poirier wrote:
>>> Hi Puranjay,
> 
> Removed Puranjay (as he is no longer with TI) and adding Danish.
> 
> Regards,
> Kishon
>>>
>>> On Tue, Jun 07, 2022 at 10:26:46AM +0530, Puranjay Mohan wrote:
>>>> From: Tero Kristo <t-kristo@ti.com>
>>>>
>>>> Add two new APIs, pru_rproc_get() and pru_rproc_put(), to the PRU
>>>> driver to allow client drivers to acquire and release the remoteproc
>>>> device associated with a PRU core. The PRU cores are treated as
>>>> resources with only one client owning it at a time.
>>>>
>>>> The pru_rproc_get() function returns the rproc handle corresponding
>>>> to a PRU core identified by the device tree "ti,prus" property under
>>>> the client node. The pru_rproc_put() is the complementary function
>>>> to pru_rproc_get().
>>>>
>>>> Co-developed-by: Suman Anna <s-anna@ti.com>
>>>> Signed-off-by: Suman Anna <s-anna@ti.com>
>>>> Signed-off-by: Tero Kristo <t-kristo@ti.com>
>>>> Co-developed-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
>>>> Signed-off-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
>>>> Co-developed-by: Puranjay Mohan <p-mohan@ti.com>
>>>> Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
>>>> ---
>>>>   drivers/remoteproc/pru_rproc.c | 138 +++++++++++++++++++++++++++++++--
>>>>   include/linux/pruss.h          |  56 +++++++++++++
>>>>   2 files changed, 189 insertions(+), 5 deletions(-)
>>>>   create mode 100644 include/linux/pruss.h
>>>>
>>>> diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
>>>> index 1777a01fa84e..7a35b400287a 100644
>>>> --- a/drivers/remoteproc/pru_rproc.c
>>>> +++ b/drivers/remoteproc/pru_rproc.c
>>>> @@ -2,12 +2,13 @@
>>>>   /*
>>>>    * PRU-ICSS remoteproc driver for various TI SoCs
>>>>    *
>>>> - * Copyright (C) 2014-2020 Texas Instruments Incorporated - https://www.ti.com/
>>>> + * Copyright (C) 2014-2022 Texas Instruments Incorporated - https://www.ti.com/
>>>>    *
>>>>    * Author(s):
>>>>    *	Suman Anna <s-anna@ti.com>
>>>>    *	Andrew F. Davis <afd@ti.com>
>>>>    *	Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org> for Texas Instruments
>>>> + *	Puranjay Mohan <p-mohan@ti.com>
>>>>    */
>>>>   
>>>>   #include <linux/bitops.h>
>>>> @@ -16,6 +17,7 @@
>>>>   #include <linux/module.h>
>>>>   #include <linux/of_device.h>
>>>>   #include <linux/of_irq.h>
>>>> +#include <linux/pruss.h>
>>>>   #include <linux/pruss_driver.h>
>>>>   #include <linux/remoteproc.h>
>>>>   
>>>> @@ -111,6 +113,8 @@ struct pru_private_data {
>>>>    * @rproc: remoteproc pointer for this PRU core
>>>>    * @data: PRU core specific data
>>>>    * @mem_regions: data for each of the PRU memory regions
>>>> + * @client_np: client device node
>>>> + * @lock: mutex to protect client usage
>>>>    * @fw_name: name of firmware image used during loading
>>>>    * @mapped_irq: virtual interrupt numbers of created fw specific mapping
>>>>    * @pru_interrupt_map: pointer to interrupt mapping description (firmware)
>>>> @@ -126,6 +130,8 @@ struct pru_rproc {
>>>>   	struct rproc *rproc;
>>>>   	const struct pru_private_data *data;
>>>>   	struct pruss_mem_region mem_regions[PRU_IOMEM_MAX];
>>>> +	struct device_node *client_np;
>>>> +	struct mutex lock; /* client access lock */
>>>>   	const char *fw_name;
>>>>   	unsigned int *mapped_irq;
>>>>   	struct pru_irq_rsc *pru_interrupt_map;
>>>> @@ -146,6 +152,125 @@ void pru_control_write_reg(struct pru_rproc *pru, unsigned int reg, u32 val)
>>>>   	writel_relaxed(val, pru->mem_regions[PRU_IOMEM_CTRL].va + reg);
>>>>   }
>>>>   
>>>> +static struct rproc *__pru_rproc_get(struct device_node *np, int index)
>>>> +{
>>>> +	struct device_node *rproc_np = NULL;
>>>> +	struct platform_device *pdev;
>>>> +	struct rproc *rproc;
>>>> +
>>>> +	rproc_np = of_parse_phandle(np, "ti,prus", index);
>>>> +	if (!rproc_np || !of_device_is_available(rproc_np))
>>>> +		return ERR_PTR(-ENODEV);
>>>> +
>>>> +	pdev = of_find_device_by_node(rproc_np);
>>>> +	of_node_put(rproc_np);
>>>> +
>>>> +	if (!pdev || !(&pdev->dev) || !((&pdev->dev)->driver))
>>>> +		/* probably PRU not yet probed */
>>>> +		return ERR_PTR(-EPROBE_DEFER);
>>>> +
>>>> +	/* make sure it is PRU rproc */
>>>> +	if (!is_pru_rproc(&pdev->dev)) {
>>>> +		put_device(&pdev->dev);
>>>> +		return ERR_PTR(-ENODEV);
>>>> +	}
>>>> +
>>>> +	rproc = platform_get_drvdata(pdev);
>>>> +	put_device(&pdev->dev);
>>>> +	if (!rproc)
>>>> +		return ERR_PTR(-EPROBE_DEFER);
>>>> +
>>>> +	get_device(&rproc->dev);
>>>> +
>>>> +	return rproc;
>>>> +}
>>>> +
>>>> +/**
>>>> + * pru_rproc_get() - get the PRU rproc instance from a device node
>>>> + * @np: the user/client device node
>>>> + * @index: index to use for the ti,prus property
>>>> + * @pru_id: optional pointer to return the PRU remoteproc processor id
>>>> + *
>>>> + * This function looks through a client device node's "ti,prus" property at
>>>> + * index @index and returns the rproc handle for a valid PRU remote processor if
>>>> + * found. The function allows only one user to own the PRU rproc resource at a
>>>> + * time. Caller must call pru_rproc_put() when done with using the rproc, not
>>>> + * required if the function returns a failure.
>>>> + *
>>>> + * When optional @pru_id pointer is passed the PRU remoteproc processor id is
>>>> + * returned.
>>>> + *
>>>> + * Return: rproc handle on success, and an ERR_PTR on failure using one
>>>> + * of the following error values
>>>> + *    -ENODEV if device is not found
>>>> + *    -EBUSY if PRU is already acquired by anyone
>>>> + *    -EPROBE_DEFER is PRU device is not probed yet
>>>> + */
>>>> +struct rproc *pru_rproc_get(struct device_node *np, int index,
>>>> +			    enum pruss_pru_id *pru_id)
>>>> +{
>>>> +	struct rproc *rproc;
>>>> +	struct pru_rproc *pru;
>>>> +	struct device *dev;
>>>> +
>>>> +	try_module_get(THIS_MODULE);
>>>
>>> There should be a module_put() in pru_rproc_put()...
>>
>> ... and in the error path of this function.
>>
>>>
>>> More comments to come tomorrow.  I'm especially worried about this API racing
>>> with a remote processor being removed or detached.
>>>
>>
>> Looking at what is done in wkup_m3_ipc_probe(), it should be possible to call
>> rproc_get_by_handle() here and that would make sure the remote processor doesn't
>> go away before the end of the function.
>>
>> More comments to come...

It is possible to call rproc_get_by_handle() here instead of 
__pru_get_proc(), but that would not provide multiple functionality.

The API rproc_get_by_handle() returns rproc handle on success, and NULL 
on failure where as __pru_get_proc() returns ERR_PTR on failure which 
provides multiple functionality and opportunity for us to distinguish 
between multiple errors.

So we have these three options.

1. If we're using the API rproc_get_by_handle() and we want the multiple 
ERR_PTR on failure then we will need to change the API 
rproc_get_by_handle() and also all the functions that uses 
rproc_get_by_handle().

2. Keep the API rproc_get_by_handle() as it is. That will restrict us 
from using multiple ERR_PTR on different kinds of error.

3. Instead of using rproc_get_by_handle(), keep using __pru_get_proc(). 
This will make sure we have the proper ERR_PTR to retrun for different 
kinds of errors.

Please let me know which option to continue with.

Thanks,
Danish


>>
>>
>>> Thanks,
>>> Mathieu
>>>
>>>> +
>>>> +	rproc = __pru_rproc_get(np, index);
>>>> +	if (IS_ERR(rproc))
>>>> +		return rproc;
>>>> +
>>>> +	pru = rproc->priv;
>>>> +	dev = &rproc->dev;
>>>> +
>>>> +	mutex_lock(&pru->lock);
>>>> +
>>>> +	if (pru->client_np) {
>>>> +		mutex_unlock(&pru->lock);
>>>> +		put_device(dev);
>>>> +		return ERR_PTR(-EBUSY);
>>>> +	}
>>>> +
>>>> +	pru->client_np = np;
>>>> +
>>>> +	mutex_unlock(&pru->lock);
>>>> +
>>>> +	if (pru_id)
>>>> +		*pru_id = pru->id;
>>>> +
>>>> +	return rproc;
>>>> +}
>>>> +EXPORT_SYMBOL_GPL(pru_rproc_get);
>>>> +
>>>> +/**
>>>> + * pru_rproc_put() - release the PRU rproc resource
>>>> + * @rproc: the rproc resource to release
>>>> + *
>>>> + * Releases the PRU rproc resource and makes it available to other
>>>> + * users.
>>>> + */
>>>> +void pru_rproc_put(struct rproc *rproc)
>>>> +{
>>>> +	struct pru_rproc *pru;
>>>> +
>>>> +	if (IS_ERR_OR_NULL(rproc) || !is_pru_rproc(rproc->dev.parent))
>>>> +		return;
>>>> +
>>>> +	pru = rproc->priv;
>>>> +
>>>> +	mutex_lock(&pru->lock);
>>>> +
>>>> +	if (!pru->client_np) {
>>>> +		mutex_unlock(&pru->lock);
>>>> +		return;
>>>> +	}
>>>> +
>>>> +	pru->client_np = NULL;
>>>> +	mutex_unlock(&pru->lock);
>>>> +
>>>> +	put_device(&rproc->dev);
>>>> +}
>>>> +EXPORT_SYMBOL_GPL(pru_rproc_put);
>>>> +
>>>>   static inline u32 pru_debug_read_reg(struct pru_rproc *pru, unsigned int reg)
>>>>   {
>>>>   	return readl_relaxed(pru->mem_regions[PRU_IOMEM_DEBUG].va + reg);
>>>> @@ -438,7 +563,7 @@ static void *pru_d_da_to_va(struct pru_rproc *pru, u32 da, size_t len)
>>>>   	dram0 = pruss->mem_regions[PRUSS_MEM_DRAM0];
>>>>   	dram1 = pruss->mem_regions[PRUSS_MEM_DRAM1];
>>>>   	/* PRU1 has its local RAM addresses reversed */
>>>> -	if (pru->id == 1)
>>>> +	if (pru->id == PRUSS_PRU1)
>>>>   		swap(dram0, dram1);
>>>>   	shrd_ram = pruss->mem_regions[PRUSS_MEM_SHRD_RAM2];
>>>>   
>>>> @@ -747,14 +872,14 @@ static int pru_rproc_set_id(struct pru_rproc *pru)
>>>>   	case RTU0_IRAM_ADDR_MASK:
>>>>   		fallthrough;
>>>>   	case PRU0_IRAM_ADDR_MASK:
>>>> -		pru->id = 0;
>>>> +		pru->id = PRUSS_PRU0;
>>>>   		break;
>>>>   	case TX_PRU1_IRAM_ADDR_MASK:
>>>>   		fallthrough;
>>>>   	case RTU1_IRAM_ADDR_MASK:
>>>>   		fallthrough;
>>>>   	case PRU1_IRAM_ADDR_MASK:
>>>> -		pru->id = 1;
>>>> +		pru->id = PRUSS_PRU1;
>>>>   		break;
>>>>   	default:
>>>>   		ret = -EINVAL;
>>>> @@ -816,6 +941,8 @@ static int pru_rproc_probe(struct platform_device *pdev)
>>>>   	pru->pruss = platform_get_drvdata(ppdev);
>>>>   	pru->rproc = rproc;
>>>>   	pru->fw_name = fw_name;
>>>> +	pru->client_np = NULL;
>>>> +	mutex_init(&pru->lock);
>>>>   
>>>>   	for (i = 0; i < ARRAY_SIZE(mem_names); i++) {
>>>>   		res = platform_get_resource_byname(pdev, IORESOURCE_MEM,
>>>> @@ -903,7 +1030,7 @@ MODULE_DEVICE_TABLE(of, pru_rproc_match);
>>>>   
>>>>   static struct platform_driver pru_rproc_driver = {
>>>>   	.driver = {
>>>> -		.name   = "pru-rproc",
>>>> +		.name   = PRU_RPROC_DRVNAME,
>>>>   		.of_match_table = pru_rproc_match,
>>>>   		.suppress_bind_attrs = true,
>>>>   	},
>>>> @@ -915,5 +1042,6 @@ module_platform_driver(pru_rproc_driver);
>>>>   MODULE_AUTHOR("Suman Anna <s-anna@ti.com>");
>>>>   MODULE_AUTHOR("Andrew F. Davis <afd@ti.com>");
>>>>   MODULE_AUTHOR("Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>");
>>>> +MODULE_AUTHOR("Puranjay Mohan <p-mohan@ti.com>");
>>>>   MODULE_DESCRIPTION("PRU-ICSS Remote Processor Driver");
>>>>   MODULE_LICENSE("GPL v2");
>>>> diff --git a/include/linux/pruss.h b/include/linux/pruss.h
>>>> new file mode 100644
>>>> index 000000000000..fdc719b43db0
>>>> --- /dev/null
>>>> +++ b/include/linux/pruss.h
>>>> @@ -0,0 +1,56 @@
>>>> +/* SPDX-License-Identifier: GPL-2.0-only */
>>>> +/**
>>>> + * PRU-ICSS Subsystem user interfaces
>>>> + *
>>>> + * Copyright (C) 2015-2022 Texas Instruments Incorporated - http://www.ti.com
>>>> + *	Suman Anna <s-anna@ti.com>
>>>> + */
>>>> +
>>>> +#ifndef __LINUX_PRUSS_H
>>>> +#define __LINUX_PRUSS_H
>>>> +
>>>> +#include <linux/device.h>
>>>> +#include <linux/types.h>
>>>> +
>>>> +#define PRU_RPROC_DRVNAME "pru-rproc"
>>>> +
>>>> +/*
>>>> + * enum pruss_pru_id - PRU core identifiers
>>>> + */
>>>> +enum pruss_pru_id {
>>>> +	PRUSS_PRU0 = 0,
>>>> +	PRUSS_PRU1,
>>>> +	PRUSS_NUM_PRUS,
>>>> +};
>>>> +
>>>> +struct device_node;
>>>> +
>>>> +#if IS_ENABLED(CONFIG_PRU_REMOTEPROC)
>>>> +
>>>> +struct rproc *pru_rproc_get(struct device_node *np, int index,
>>>> +			    enum pruss_pru_id *pru_id);
>>>> +void pru_rproc_put(struct rproc *rproc);
>>>> +
>>>> +#else
>>>> +
>>>> +static inline struct rproc *
>>>> +pru_rproc_get(struct device_node *np, int index, enum pruss_pru_id *pru_id)
>>>> +{
>>>> +	return ERR_PTR(-EOPNOTSUPP);
>>>> +}
>>>> +
>>>> +static inline void pru_rproc_put(struct rproc *rproc) { }
>>>> +
>>>> +#endif /* CONFIG_PRU_REMOTEPROC */
>>>> +
>>>> +static inline bool is_pru_rproc(struct device *dev)
>>>> +{
>>>> +	const char *drv_name = dev_driver_string(dev);
>>>> +
>>>> +	if (strncmp(drv_name, PRU_RPROC_DRVNAME, sizeof(PRU_RPROC_DRVNAME)))
>>>> +		return false;
>>>> +
>>>> +	return true;
>>>> +}
>>>> +
>>>> +#endif /* __LINUX_PRUSS_H */
>>>> -- 
>>>> 2.17.1
>>>>

Re: [PATCH v5 2/6] remoteproc: pru: Add APIs to get and put the PRU cores

[Mathieu Poirier]

On Fri, Sep 02, 2022 at 03:09:04PM +0530, Md Danish Anwar wrote:
> Hi Mathieu,
> 
> 
> On 15/07/22 11:52, Kishon Vijay Abraham I wrote:
> > +Danish
> > 
> > Hi Mathieu,
> > 
> > On 14/07/22 22:50, Mathieu Poirier wrote:
> > > On Tue, Jul 12, 2022 at 11:59:49AM -0600, Mathieu Poirier wrote:
> > > > Hi Puranjay,
> > 
> > Removed Puranjay (as he is no longer with TI) and adding Danish.
> > 
> > Regards,
> > Kishon
> > > > 
> > > > On Tue, Jun 07, 2022 at 10:26:46AM +0530, Puranjay Mohan wrote:
> > > > > From: Tero Kristo <t-kristo@ti.com>
> > > > > 
> > > > > Add two new APIs, pru_rproc_get() and pru_rproc_put(), to the PRU
> > > > > driver to allow client drivers to acquire and release the remoteproc
> > > > > device associated with a PRU core. The PRU cores are treated as
> > > > > resources with only one client owning it at a time.
> > > > > 
> > > > > The pru_rproc_get() function returns the rproc handle corresponding
> > > > > to a PRU core identified by the device tree "ti,prus" property under
> > > > > the client node. The pru_rproc_put() is the complementary function
> > > > > to pru_rproc_get().
> > > > > 
> > > > > Co-developed-by: Suman Anna <s-anna@ti.com>
> > > > > Signed-off-by: Suman Anna <s-anna@ti.com>
> > > > > Signed-off-by: Tero Kristo <t-kristo@ti.com>
> > > > > Co-developed-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
> > > > > Signed-off-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
> > > > > Co-developed-by: Puranjay Mohan <p-mohan@ti.com>
> > > > > Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
> > > > > ---
> > > > >   drivers/remoteproc/pru_rproc.c | 138 +++++++++++++++++++++++++++++++--
> > > > >   include/linux/pruss.h          |  56 +++++++++++++
> > > > >   2 files changed, 189 insertions(+), 5 deletions(-)
> > > > >   create mode 100644 include/linux/pruss.h
> > > > > 
> > > > > diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
> > > > > index 1777a01fa84e..7a35b400287a 100644
> > > > > --- a/drivers/remoteproc/pru_rproc.c
> > > > > +++ b/drivers/remoteproc/pru_rproc.c
> > > > > @@ -2,12 +2,13 @@
> > > > >   /*
> > > > >    * PRU-ICSS remoteproc driver for various TI SoCs
> > > > >    *
> > > > > - * Copyright (C) 2014-2020 Texas Instruments Incorporated - https://www.ti.com/
> > > > > + * Copyright (C) 2014-2022 Texas Instruments Incorporated - https://www.ti.com/
> > > > >    *
> > > > >    * Author(s):
> > > > >    *	Suman Anna <s-anna@ti.com>
> > > > >    *	Andrew F. Davis <afd@ti.com>
> > > > >    *	Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org> for Texas Instruments
> > > > > + *	Puranjay Mohan <p-mohan@ti.com>
> > > > >    */
> > > > >   #include <linux/bitops.h>
> > > > > @@ -16,6 +17,7 @@
> > > > >   #include <linux/module.h>
> > > > >   #include <linux/of_device.h>
> > > > >   #include <linux/of_irq.h>
> > > > > +#include <linux/pruss.h>
> > > > >   #include <linux/pruss_driver.h>
> > > > >   #include <linux/remoteproc.h>
> > > > > @@ -111,6 +113,8 @@ struct pru_private_data {
> > > > >    * @rproc: remoteproc pointer for this PRU core
> > > > >    * @data: PRU core specific data
> > > > >    * @mem_regions: data for each of the PRU memory regions
> > > > > + * @client_np: client device node
> > > > > + * @lock: mutex to protect client usage
> > > > >    * @fw_name: name of firmware image used during loading
> > > > >    * @mapped_irq: virtual interrupt numbers of created fw specific mapping
> > > > >    * @pru_interrupt_map: pointer to interrupt mapping description (firmware)
> > > > > @@ -126,6 +130,8 @@ struct pru_rproc {
> > > > >   	struct rproc *rproc;
> > > > >   	const struct pru_private_data *data;
> > > > >   	struct pruss_mem_region mem_regions[PRU_IOMEM_MAX];
> > > > > +	struct device_node *client_np;
> > > > > +	struct mutex lock; /* client access lock */
> > > > >   	const char *fw_name;
> > > > >   	unsigned int *mapped_irq;
> > > > >   	struct pru_irq_rsc *pru_interrupt_map;
> > > > > @@ -146,6 +152,125 @@ void pru_control_write_reg(struct pru_rproc *pru, unsigned int reg, u32 val)
> > > > >   	writel_relaxed(val, pru->mem_regions[PRU_IOMEM_CTRL].va + reg);
> > > > >   }
> > > > > +static struct rproc *__pru_rproc_get(struct device_node *np, int index)
> > > > > +{
> > > > > +	struct device_node *rproc_np = NULL;
> > > > > +	struct platform_device *pdev;
> > > > > +	struct rproc *rproc;
> > > > > +
> > > > > +	rproc_np = of_parse_phandle(np, "ti,prus", index);
> > > > > +	if (!rproc_np || !of_device_is_available(rproc_np))
> > > > > +		return ERR_PTR(-ENODEV);
> > > > > +
> > > > > +	pdev = of_find_device_by_node(rproc_np);
> > > > > +	of_node_put(rproc_np);
> > > > > +
> > > > > +	if (!pdev || !(&pdev->dev) || !((&pdev->dev)->driver))
> > > > > +		/* probably PRU not yet probed */
> > > > > +		return ERR_PTR(-EPROBE_DEFER);
> > > > > +
> > > > > +	/* make sure it is PRU rproc */
> > > > > +	if (!is_pru_rproc(&pdev->dev)) {
> > > > > +		put_device(&pdev->dev);
> > > > > +		return ERR_PTR(-ENODEV);
> > > > > +	}
> > > > > +
> > > > > +	rproc = platform_get_drvdata(pdev);
> > > > > +	put_device(&pdev->dev);
> > > > > +	if (!rproc)
> > > > > +		return ERR_PTR(-EPROBE_DEFER);
> > > > > +
> > > > > +	get_device(&rproc->dev);
> > > > > +
> > > > > +	return rproc;
> > > > > +}
> > > > > +
> > > > > +/**
> > > > > + * pru_rproc_get() - get the PRU rproc instance from a device node
> > > > > + * @np: the user/client device node
> > > > > + * @index: index to use for the ti,prus property
> > > > > + * @pru_id: optional pointer to return the PRU remoteproc processor id
> > > > > + *
> > > > > + * This function looks through a client device node's "ti,prus" property at
> > > > > + * index @index and returns the rproc handle for a valid PRU remote processor if
> > > > > + * found. The function allows only one user to own the PRU rproc resource at a
> > > > > + * time. Caller must call pru_rproc_put() when done with using the rproc, not
> > > > > + * required if the function returns a failure.
> > > > > + *
> > > > > + * When optional @pru_id pointer is passed the PRU remoteproc processor id is
> > > > > + * returned.
> > > > > + *
> > > > > + * Return: rproc handle on success, and an ERR_PTR on failure using one
> > > > > + * of the following error values
> > > > > + *    -ENODEV if device is not found
> > > > > + *    -EBUSY if PRU is already acquired by anyone
> > > > > + *    -EPROBE_DEFER is PRU device is not probed yet
> > > > > + */
> > > > > +struct rproc *pru_rproc_get(struct device_node *np, int index,
> > > > > +			    enum pruss_pru_id *pru_id)
> > > > > +{
> > > > > +	struct rproc *rproc;
> > > > > +	struct pru_rproc *pru;
> > > > > +	struct device *dev;
> > > > > +
> > > > > +	try_module_get(THIS_MODULE);
> > > > 
> > > > There should be a module_put() in pru_rproc_put()...
> > > 
> > > ... and in the error path of this function.
> > > 
> > > > 
> > > > More comments to come tomorrow.  I'm especially worried about this API racing
> > > > with a remote processor being removed or detached.
> > > > 
> > > 
> > > Looking at what is done in wkup_m3_ipc_probe(), it should be possible to call
> > > rproc_get_by_handle() here and that would make sure the remote processor doesn't
> > > go away before the end of the function.
> > > 
> > > More comments to come...
> 
> It is possible to call rproc_get_by_handle() here instead of
> __pru_get_proc(), but that would not provide multiple functionality.
> 
> The API rproc_get_by_handle() returns rproc handle on success, and NULL on
> failure where as __pru_get_proc() returns ERR_PTR on failure which provides
> multiple functionality and opportunity for us to distinguish between
> multiple errors.
> 
> So we have these three options.
> 
> 1. If we're using the API rproc_get_by_handle() and we want the multiple
> ERR_PTR on failure then we will need to change the API rproc_get_by_handle()
> and also all the functions that uses rproc_get_by_handle().
>

Not optimal.

> 2. Keep the API rproc_get_by_handle() as it is. That will restrict us from
> using multiple ERR_PTR on different kinds of error.
> 

Not optimal.

> 3. Instead of using rproc_get_by_handle(), keep using __pru_get_proc(). This
> will make sure we have the proper ERR_PTR to retrun for different kinds of
> errors.
> 

Unacceptable for the reason I already stated.  

> Please let me know which option to continue with.

I suggest building a wrapper that does everything you want around rproc_get_by_phandle().

> 
> Thanks,
> Danish
> 
> 
> > > 
> > > 
> > > > Thanks,
> > > > Mathieu
> > > > 
> > > > > +
> > > > > +	rproc = __pru_rproc_get(np, index);
> > > > > +	if (IS_ERR(rproc))
> > > > > +		return rproc;
> > > > > +
> > > > > +	pru = rproc->priv;
> > > > > +	dev = &rproc->dev;
> > > > > +
> > > > > +	mutex_lock(&pru->lock);
> > > > > +
> > > > > +	if (pru->client_np) {
> > > > > +		mutex_unlock(&pru->lock);
> > > > > +		put_device(dev);
> > > > > +		return ERR_PTR(-EBUSY);
> > > > > +	}
> > > > > +
> > > > > +	pru->client_np = np;
> > > > > +
> > > > > +	mutex_unlock(&pru->lock);
> > > > > +
> > > > > +	if (pru_id)
> > > > > +		*pru_id = pru->id;
> > > > > +
> > > > > +	return rproc;
> > > > > +}
> > > > > +EXPORT_SYMBOL_GPL(pru_rproc_get);
> > > > > +
> > > > > +/**
> > > > > + * pru_rproc_put() - release the PRU rproc resource
> > > > > + * @rproc: the rproc resource to release
> > > > > + *
> > > > > + * Releases the PRU rproc resource and makes it available to other
> > > > > + * users.
> > > > > + */
> > > > > +void pru_rproc_put(struct rproc *rproc)
> > > > > +{
> > > > > +	struct pru_rproc *pru;
> > > > > +
> > > > > +	if (IS_ERR_OR_NULL(rproc) || !is_pru_rproc(rproc->dev.parent))
> > > > > +		return;
> > > > > +
> > > > > +	pru = rproc->priv;
> > > > > +
> > > > > +	mutex_lock(&pru->lock);
> > > > > +
> > > > > +	if (!pru->client_np) {
> > > > > +		mutex_unlock(&pru->lock);
> > > > > +		return;
> > > > > +	}
> > > > > +
> > > > > +	pru->client_np = NULL;
> > > > > +	mutex_unlock(&pru->lock);
> > > > > +
> > > > > +	put_device(&rproc->dev);
> > > > > +}
> > > > > +EXPORT_SYMBOL_GPL(pru_rproc_put);
> > > > > +
> > > > >   static inline u32 pru_debug_read_reg(struct pru_rproc *pru, unsigned int reg)
> > > > >   {
> > > > >   	return readl_relaxed(pru->mem_regions[PRU_IOMEM_DEBUG].va + reg);
> > > > > @@ -438,7 +563,7 @@ static void *pru_d_da_to_va(struct pru_rproc *pru, u32 da, size_t len)
> > > > >   	dram0 = pruss->mem_regions[PRUSS_MEM_DRAM0];
> > > > >   	dram1 = pruss->mem_regions[PRUSS_MEM_DRAM1];
> > > > >   	/* PRU1 has its local RAM addresses reversed */
> > > > > -	if (pru->id == 1)
> > > > > +	if (pru->id == PRUSS_PRU1)
> > > > >   		swap(dram0, dram1);
> > > > >   	shrd_ram = pruss->mem_regions[PRUSS_MEM_SHRD_RAM2];
> > > > > @@ -747,14 +872,14 @@ static int pru_rproc_set_id(struct pru_rproc *pru)
> > > > >   	case RTU0_IRAM_ADDR_MASK:
> > > > >   		fallthrough;
> > > > >   	case PRU0_IRAM_ADDR_MASK:
> > > > > -		pru->id = 0;
> > > > > +		pru->id = PRUSS_PRU0;
> > > > >   		break;
> > > > >   	case TX_PRU1_IRAM_ADDR_MASK:
> > > > >   		fallthrough;
> > > > >   	case RTU1_IRAM_ADDR_MASK:
> > > > >   		fallthrough;
> > > > >   	case PRU1_IRAM_ADDR_MASK:
> > > > > -		pru->id = 1;
> > > > > +		pru->id = PRUSS_PRU1;
> > > > >   		break;
> > > > >   	default:
> > > > >   		ret = -EINVAL;
> > > > > @@ -816,6 +941,8 @@ static int pru_rproc_probe(struct platform_device *pdev)
> > > > >   	pru->pruss = platform_get_drvdata(ppdev);
> > > > >   	pru->rproc = rproc;
> > > > >   	pru->fw_name = fw_name;
> > > > > +	pru->client_np = NULL;
> > > > > +	mutex_init(&pru->lock);
> > > > >   	for (i = 0; i < ARRAY_SIZE(mem_names); i++) {
> > > > >   		res = platform_get_resource_byname(pdev, IORESOURCE_MEM,
> > > > > @@ -903,7 +1030,7 @@ MODULE_DEVICE_TABLE(of, pru_rproc_match);
> > > > >   static struct platform_driver pru_rproc_driver = {
> > > > >   	.driver = {
> > > > > -		.name   = "pru-rproc",
> > > > > +		.name   = PRU_RPROC_DRVNAME,
> > > > >   		.of_match_table = pru_rproc_match,
> > > > >   		.suppress_bind_attrs = true,
> > > > >   	},
> > > > > @@ -915,5 +1042,6 @@ module_platform_driver(pru_rproc_driver);
> > > > >   MODULE_AUTHOR("Suman Anna <s-anna@ti.com>");
> > > > >   MODULE_AUTHOR("Andrew F. Davis <afd@ti.com>");
> > > > >   MODULE_AUTHOR("Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>");
> > > > > +MODULE_AUTHOR("Puranjay Mohan <p-mohan@ti.com>");
> > > > >   MODULE_DESCRIPTION("PRU-ICSS Remote Processor Driver");
> > > > >   MODULE_LICENSE("GPL v2");
> > > > > diff --git a/include/linux/pruss.h b/include/linux/pruss.h
> > > > > new file mode 100644
> > > > > index 000000000000..fdc719b43db0
> > > > > --- /dev/null
> > > > > +++ b/include/linux/pruss.h
> > > > > @@ -0,0 +1,56 @@
> > > > > +/* SPDX-License-Identifier: GPL-2.0-only */
> > > > > +/**
> > > > > + * PRU-ICSS Subsystem user interfaces
> > > > > + *
> > > > > + * Copyright (C) 2015-2022 Texas Instruments Incorporated - http://www.ti.com
> > > > > + *	Suman Anna <s-anna@ti.com>
> > > > > + */
> > > > > +
> > > > > +#ifndef __LINUX_PRUSS_H
> > > > > +#define __LINUX_PRUSS_H
> > > > > +
> > > > > +#include <linux/device.h>
> > > > > +#include <linux/types.h>
> > > > > +
> > > > > +#define PRU_RPROC_DRVNAME "pru-rproc"
> > > > > +
> > > > > +/*
> > > > > + * enum pruss_pru_id - PRU core identifiers
> > > > > + */
> > > > > +enum pruss_pru_id {
> > > > > +	PRUSS_PRU0 = 0,
> > > > > +	PRUSS_PRU1,
> > > > > +	PRUSS_NUM_PRUS,
> > > > > +};
> > > > > +
> > > > > +struct device_node;
> > > > > +
> > > > > +#if IS_ENABLED(CONFIG_PRU_REMOTEPROC)
> > > > > +
> > > > > +struct rproc *pru_rproc_get(struct device_node *np, int index,
> > > > > +			    enum pruss_pru_id *pru_id);
> > > > > +void pru_rproc_put(struct rproc *rproc);
> > > > > +
> > > > > +#else
> > > > > +
> > > > > +static inline struct rproc *
> > > > > +pru_rproc_get(struct device_node *np, int index, enum pruss_pru_id *pru_id)
> > > > > +{
> > > > > +	return ERR_PTR(-EOPNOTSUPP);
> > > > > +}
> > > > > +
> > > > > +static inline void pru_rproc_put(struct rproc *rproc) { }
> > > > > +
> > > > > +#endif /* CONFIG_PRU_REMOTEPROC */
> > > > > +
> > > > > +static inline bool is_pru_rproc(struct device *dev)
> > > > > +{
> > > > > +	const char *drv_name = dev_driver_string(dev);
> > > > > +
> > > > > +	if (strncmp(drv_name, PRU_RPROC_DRVNAME, sizeof(PRU_RPROC_DRVNAME)))
> > > > > +		return false;
> > > > > +
> > > > > +	return true;
> > > > > +}
> > > > > +
> > > > > +#endif /* __LINUX_PRUSS_H */
> > > > > -- 
> > > > > 2.17.1
> > > > > 

Re: [PATCH v5 2/6] remoteproc: pru: Add APIs to get and put the PRU cores

[Md Danish Anwar]

Hi Mathieu,

On 07/09/22 00:58, Mathieu Poirier wrote:
> On Fri, Sep 02, 2022 at 03:09:04PM +0530, Md Danish Anwar wrote:
>> Hi Mathieu,
>>
>>
>> On 15/07/22 11:52, Kishon Vijay Abraham I wrote:
>>> +Danish
>>>
>>> Hi Mathieu,
>>>
>>> On 14/07/22 22:50, Mathieu Poirier wrote:
>>>> On Tue, Jul 12, 2022 at 11:59:49AM -0600, Mathieu Poirier wrote:
>>>>> Hi Puranjay,
>>>
>>> Removed Puranjay (as he is no longer with TI) and adding Danish.
>>>
>>> Regards,
>>> Kishon
>>>>>
>>>>> On Tue, Jun 07, 2022 at 10:26:46AM +0530, Puranjay Mohan wrote:
>>>>>> From: Tero Kristo <t-kristo@ti.com>
>>>>>>
>>>>>> Add two new APIs, pru_rproc_get() and pru_rproc_put(), to the PRU
>>>>>> driver to allow client drivers to acquire and release the remoteproc
>>>>>> device associated with a PRU core. The PRU cores are treated as
>>>>>> resources with only one client owning it at a time.
>>>>>>
>>>>>> The pru_rproc_get() function returns the rproc handle corresponding
>>>>>> to a PRU core identified by the device tree "ti,prus" property under
>>>>>> the client node. The pru_rproc_put() is the complementary function
>>>>>> to pru_rproc_get().
>>>>>>
>>>>>> Co-developed-by: Suman Anna <s-anna@ti.com>
>>>>>> Signed-off-by: Suman Anna <s-anna@ti.com>
>>>>>> Signed-off-by: Tero Kristo <t-kristo@ti.com>
>>>>>> Co-developed-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
>>>>>> Signed-off-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
>>>>>> Co-developed-by: Puranjay Mohan <p-mohan@ti.com>
>>>>>> Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
>>>>>> ---
>>>>>>   drivers/remoteproc/pru_rproc.c | 138 +++++++++++++++++++++++++++++++--
>>>>>>   include/linux/pruss.h          |  56 +++++++++++++
>>>>>>   2 files changed, 189 insertions(+), 5 deletions(-)
>>>>>>   create mode 100644 include/linux/pruss.h
>>>>>>
>>>>>> diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
>>>>>> index 1777a01fa84e..7a35b400287a 100644
>>>>>> --- a/drivers/remoteproc/pru_rproc.c
>>>>>> +++ b/drivers/remoteproc/pru_rproc.c
>>>>>> @@ -2,12 +2,13 @@
>>>>>>   /*
>>>>>>    * PRU-ICSS remoteproc driver for various TI SoCs
>>>>>>    *
>>>>>> - * Copyright (C) 2014-2020 Texas Instruments Incorporated - https://www.ti.com/
>>>>>> + * Copyright (C) 2014-2022 Texas Instruments Incorporated - https://www.ti.com/
>>>>>>    *
>>>>>>    * Author(s):
>>>>>>    *	Suman Anna <s-anna@ti.com>
>>>>>>    *	Andrew F. Davis <afd@ti.com>
>>>>>>    *	Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org> for Texas Instruments
>>>>>> + *	Puranjay Mohan <p-mohan@ti.com>
>>>>>>    */
>>>>>>   #include <linux/bitops.h>
>>>>>> @@ -16,6 +17,7 @@
>>>>>>   #include <linux/module.h>
>>>>>>   #include <linux/of_device.h>
>>>>>>   #include <linux/of_irq.h>
>>>>>> +#include <linux/pruss.h>
>>>>>>   #include <linux/pruss_driver.h>
>>>>>>   #include <linux/remoteproc.h>
>>>>>> @@ -111,6 +113,8 @@ struct pru_private_data {
>>>>>>    * @rproc: remoteproc pointer for this PRU core
>>>>>>    * @data: PRU core specific data
>>>>>>    * @mem_regions: data for each of the PRU memory regions
>>>>>> + * @client_np: client device node
>>>>>> + * @lock: mutex to protect client usage
>>>>>>    * @fw_name: name of firmware image used during loading
>>>>>>    * @mapped_irq: virtual interrupt numbers of created fw specific mapping
>>>>>>    * @pru_interrupt_map: pointer to interrupt mapping description (firmware)
>>>>>> @@ -126,6 +130,8 @@ struct pru_rproc {
>>>>>>   	struct rproc *rproc;
>>>>>>   	const struct pru_private_data *data;
>>>>>>   	struct pruss_mem_region mem_regions[PRU_IOMEM_MAX];
>>>>>> +	struct device_node *client_np;
>>>>>> +	struct mutex lock; /* client access lock */
>>>>>>   	const char *fw_name;
>>>>>>   	unsigned int *mapped_irq;
>>>>>>   	struct pru_irq_rsc *pru_interrupt_map;
>>>>>> @@ -146,6 +152,125 @@ void pru_control_write_reg(struct pru_rproc *pru, unsigned int reg, u32 val)
>>>>>>   	writel_relaxed(val, pru->mem_regions[PRU_IOMEM_CTRL].va + reg);
>>>>>>   }
>>>>>> +static struct rproc *__pru_rproc_get(struct device_node *np, int index)
>>>>>> +{
>>>>>> +	struct device_node *rproc_np = NULL;
>>>>>> +	struct platform_device *pdev;
>>>>>> +	struct rproc *rproc;
>>>>>> +
>>>>>> +	rproc_np = of_parse_phandle(np, "ti,prus", index);
>>>>>> +	if (!rproc_np || !of_device_is_available(rproc_np))
>>>>>> +		return ERR_PTR(-ENODEV);
>>>>>> +
>>>>>> +	pdev = of_find_device_by_node(rproc_np);
>>>>>> +	of_node_put(rproc_np);
>>>>>> +
>>>>>> +	if (!pdev || !(&pdev->dev) || !((&pdev->dev)->driver))
>>>>>> +		/* probably PRU not yet probed */
>>>>>> +		return ERR_PTR(-EPROBE_DEFER);
>>>>>> +
>>>>>> +	/* make sure it is PRU rproc */
>>>>>> +	if (!is_pru_rproc(&pdev->dev)) {
>>>>>> +		put_device(&pdev->dev);
>>>>>> +		return ERR_PTR(-ENODEV);
>>>>>> +	}
>>>>>> +
>>>>>> +	rproc = platform_get_drvdata(pdev);
>>>>>> +	put_device(&pdev->dev);
>>>>>> +	if (!rproc)
>>>>>> +		return ERR_PTR(-EPROBE_DEFER);
>>>>>> +
>>>>>> +	get_device(&rproc->dev);
>>>>>> +
>>>>>> +	return rproc;
>>>>>> +}
>>>>>> +
>>>>>> +/**
>>>>>> + * pru_rproc_get() - get the PRU rproc instance from a device node
>>>>>> + * @np: the user/client device node
>>>>>> + * @index: index to use for the ti,prus property
>>>>>> + * @pru_id: optional pointer to return the PRU remoteproc processor id
>>>>>> + *
>>>>>> + * This function looks through a client device node's "ti,prus" property at
>>>>>> + * index @index and returns the rproc handle for a valid PRU remote processor if
>>>>>> + * found. The function allows only one user to own the PRU rproc resource at a
>>>>>> + * time. Caller must call pru_rproc_put() when done with using the rproc, not
>>>>>> + * required if the function returns a failure.
>>>>>> + *
>>>>>> + * When optional @pru_id pointer is passed the PRU remoteproc processor id is
>>>>>> + * returned.
>>>>>> + *
>>>>>> + * Return: rproc handle on success, and an ERR_PTR on failure using one
>>>>>> + * of the following error values
>>>>>> + *    -ENODEV if device is not found
>>>>>> + *    -EBUSY if PRU is already acquired by anyone
>>>>>> + *    -EPROBE_DEFER is PRU device is not probed yet
>>>>>> + */
>>>>>> +struct rproc *pru_rproc_get(struct device_node *np, int index,
>>>>>> +			    enum pruss_pru_id *pru_id)
>>>>>> +{
>>>>>> +	struct rproc *rproc;
>>>>>> +	struct pru_rproc *pru;
>>>>>> +	struct device *dev;
>>>>>> +
>>>>>> +	try_module_get(THIS_MODULE);
>>>>>
>>>>> There should be a module_put() in pru_rproc_put()...
>>>>
>>>> ... and in the error path of this function.
>>>>
>>>>>
>>>>> More comments to come tomorrow.  I'm especially worried about this API racing
>>>>> with a remote processor being removed or detached.
>>>>>
>>>>
>>>> Looking at what is done in wkup_m3_ipc_probe(), it should be possible to call
>>>> rproc_get_by_handle() here and that would make sure the remote processor doesn't
>>>> go away before the end of the function.
>>>>
>>>> More comments to come...
>>
>> It is possible to call rproc_get_by_handle() here instead of
>> __pru_get_proc(), but that would not provide multiple functionality.
>>
>> The API rproc_get_by_handle() returns rproc handle on success, and NULL on
>> failure where as __pru_get_proc() returns ERR_PTR on failure which provides
>> multiple functionality and opportunity for us to distinguish between
>> multiple errors.
>>
>> So we have these three options.
>>
>> 1. If we're using the API rproc_get_by_handle() and we want the multiple
>> ERR_PTR on failure then we will need to change the API rproc_get_by_handle()
>> and also all the functions that uses rproc_get_by_handle().
>>
> 
> Not optimal.
> 
>> 2. Keep the API rproc_get_by_handle() as it is. That will restrict us from
>> using multiple ERR_PTR on different kinds of error.
>>
> 
> Not optimal.
> 
>> 3. Instead of using rproc_get_by_handle(), keep using __pru_get_proc(). This
>> will make sure we have the proper ERR_PTR to retrun for different kinds of
>> errors.
>>
> 
> Unacceptable for the reason I already stated.  
> 
>> Please let me know which option to continue with.
> 
> I suggest building a wrapper that does everything you want around rproc_get_by_phandle().
> 

Sure, I'll do that. I'll share this change in v6 of this patch series.

>>
>> Thanks,
>> Danish
>>
>>
>>>>
>>>>
>>>>> Thanks,
>>>>> Mathieu
>>>>>
>>>>>> +
>>>>>> +	rproc = __pru_rproc_get(np, index);
>>>>>> +	if (IS_ERR(rproc))
>>>>>> +		return rproc;
>>>>>> +
>>>>>> +	pru = rproc->priv;
>>>>>> +	dev = &rproc->dev;
>>>>>> +
>>>>>> +	mutex_lock(&pru->lock);
>>>>>> +
>>>>>> +	if (pru->client_np) {
>>>>>> +		mutex_unlock(&pru->lock);
>>>>>> +		put_device(dev);
>>>>>> +		return ERR_PTR(-EBUSY);
>>>>>> +	}
>>>>>> +
>>>>>> +	pru->client_np = np;
>>>>>> +
>>>>>> +	mutex_unlock(&pru->lock);
>>>>>> +
>>>>>> +	if (pru_id)
>>>>>> +		*pru_id = pru->id;
>>>>>> +
>>>>>> +	return rproc;
>>>>>> +}
>>>>>> +EXPORT_SYMBOL_GPL(pru_rproc_get);
>>>>>> +
>>>>>> +/**
>>>>>> + * pru_rproc_put() - release the PRU rproc resource
>>>>>> + * @rproc: the rproc resource to release
>>>>>> + *
>>>>>> + * Releases the PRU rproc resource and makes it available to other
>>>>>> + * users.
>>>>>> + */
>>>>>> +void pru_rproc_put(struct rproc *rproc)
>>>>>> +{
>>>>>> +	struct pru_rproc *pru;
>>>>>> +
>>>>>> +	if (IS_ERR_OR_NULL(rproc) || !is_pru_rproc(rproc->dev.parent))
>>>>>> +		return;
>>>>>> +
>>>>>> +	pru = rproc->priv;
>>>>>> +
>>>>>> +	mutex_lock(&pru->lock);
>>>>>> +
>>>>>> +	if (!pru->client_np) {
>>>>>> +		mutex_unlock(&pru->lock);
>>>>>> +		return;
>>>>>> +	}
>>>>>> +
>>>>>> +	pru->client_np = NULL;
>>>>>> +	mutex_unlock(&pru->lock);
>>>>>> +
>>>>>> +	put_device(&rproc->dev);
>>>>>> +}
>>>>>> +EXPORT_SYMBOL_GPL(pru_rproc_put);
>>>>>> +
>>>>>>   static inline u32 pru_debug_read_reg(struct pru_rproc *pru, unsigned int reg)
>>>>>>   {
>>>>>>   	return readl_relaxed(pru->mem_regions[PRU_IOMEM_DEBUG].va + reg);
>>>>>> @@ -438,7 +563,7 @@ static void *pru_d_da_to_va(struct pru_rproc *pru, u32 da, size_t len)
>>>>>>   	dram0 = pruss->mem_regions[PRUSS_MEM_DRAM0];
>>>>>>   	dram1 = pruss->mem_regions[PRUSS_MEM_DRAM1];
>>>>>>   	/* PRU1 has its local RAM addresses reversed */
>>>>>> -	if (pru->id == 1)
>>>>>> +	if (pru->id == PRUSS_PRU1)
>>>>>>   		swap(dram0, dram1);
>>>>>>   	shrd_ram = pruss->mem_regions[PRUSS_MEM_SHRD_RAM2];
>>>>>> @@ -747,14 +872,14 @@ static int pru_rproc_set_id(struct pru_rproc *pru)
>>>>>>   	case RTU0_IRAM_ADDR_MASK:
>>>>>>   		fallthrough;
>>>>>>   	case PRU0_IRAM_ADDR_MASK:
>>>>>> -		pru->id = 0;
>>>>>> +		pru->id = PRUSS_PRU0;
>>>>>>   		break;
>>>>>>   	case TX_PRU1_IRAM_ADDR_MASK:
>>>>>>   		fallthrough;
>>>>>>   	case RTU1_IRAM_ADDR_MASK:
>>>>>>   		fallthrough;
>>>>>>   	case PRU1_IRAM_ADDR_MASK:
>>>>>> -		pru->id = 1;
>>>>>> +		pru->id = PRUSS_PRU1;
>>>>>>   		break;
>>>>>>   	default:
>>>>>>   		ret = -EINVAL;
>>>>>> @@ -816,6 +941,8 @@ static int pru_rproc_probe(struct platform_device *pdev)
>>>>>>   	pru->pruss = platform_get_drvdata(ppdev);
>>>>>>   	pru->rproc = rproc;
>>>>>>   	pru->fw_name = fw_name;
>>>>>> +	pru->client_np = NULL;
>>>>>> +	mutex_init(&pru->lock);
>>>>>>   	for (i = 0; i < ARRAY_SIZE(mem_names); i++) {
>>>>>>   		res = platform_get_resource_byname(pdev, IORESOURCE_MEM,
>>>>>> @@ -903,7 +1030,7 @@ MODULE_DEVICE_TABLE(of, pru_rproc_match);
>>>>>>   static struct platform_driver pru_rproc_driver = {
>>>>>>   	.driver = {
>>>>>> -		.name   = "pru-rproc",
>>>>>> +		.name   = PRU_RPROC_DRVNAME,
>>>>>>   		.of_match_table = pru_rproc_match,
>>>>>>   		.suppress_bind_attrs = true,
>>>>>>   	},
>>>>>> @@ -915,5 +1042,6 @@ module_platform_driver(pru_rproc_driver);
>>>>>>   MODULE_AUTHOR("Suman Anna <s-anna@ti.com>");
>>>>>>   MODULE_AUTHOR("Andrew F. Davis <afd@ti.com>");
>>>>>>   MODULE_AUTHOR("Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>");
>>>>>> +MODULE_AUTHOR("Puranjay Mohan <p-mohan@ti.com>");
>>>>>>   MODULE_DESCRIPTION("PRU-ICSS Remote Processor Driver");
>>>>>>   MODULE_LICENSE("GPL v2");
>>>>>> diff --git a/include/linux/pruss.h b/include/linux/pruss.h
>>>>>> new file mode 100644
>>>>>> index 000000000000..fdc719b43db0
>>>>>> --- /dev/null
>>>>>> +++ b/include/linux/pruss.h
>>>>>> @@ -0,0 +1,56 @@
>>>>>> +/* SPDX-License-Identifier: GPL-2.0-only */
>>>>>> +/**
>>>>>> + * PRU-ICSS Subsystem user interfaces
>>>>>> + *
>>>>>> + * Copyright (C) 2015-2022 Texas Instruments Incorporated - http://www.ti.com
>>>>>> + *	Suman Anna <s-anna@ti.com>
>>>>>> + */
>>>>>> +
>>>>>> +#ifndef __LINUX_PRUSS_H
>>>>>> +#define __LINUX_PRUSS_H
>>>>>> +
>>>>>> +#include <linux/device.h>
>>>>>> +#include <linux/types.h>
>>>>>> +
>>>>>> +#define PRU_RPROC_DRVNAME "pru-rproc"
>>>>>> +
>>>>>> +/*
>>>>>> + * enum pruss_pru_id - PRU core identifiers
>>>>>> + */
>>>>>> +enum pruss_pru_id {
>>>>>> +	PRUSS_PRU0 = 0,
>>>>>> +	PRUSS_PRU1,
>>>>>> +	PRUSS_NUM_PRUS,
>>>>>> +};
>>>>>> +
>>>>>> +struct device_node;
>>>>>> +
>>>>>> +#if IS_ENABLED(CONFIG_PRU_REMOTEPROC)
>>>>>> +
>>>>>> +struct rproc *pru_rproc_get(struct device_node *np, int index,
>>>>>> +			    enum pruss_pru_id *pru_id);
>>>>>> +void pru_rproc_put(struct rproc *rproc);
>>>>>> +
>>>>>> +#else
>>>>>> +
>>>>>> +static inline struct rproc *
>>>>>> +pru_rproc_get(struct device_node *np, int index, enum pruss_pru_id *pru_id)
>>>>>> +{
>>>>>> +	return ERR_PTR(-EOPNOTSUPP);
>>>>>> +}
>>>>>> +
>>>>>> +static inline void pru_rproc_put(struct rproc *rproc) { }
>>>>>> +
>>>>>> +#endif /* CONFIG_PRU_REMOTEPROC */
>>>>>> +
>>>>>> +static inline bool is_pru_rproc(struct device *dev)
>>>>>> +{
>>>>>> +	const char *drv_name = dev_driver_string(dev);
>>>>>> +
>>>>>> +	if (strncmp(drv_name, PRU_RPROC_DRVNAME, sizeof(PRU_RPROC_DRVNAME)))
>>>>>> +		return false;
>>>>>> +
>>>>>> +	return true;
>>>>>> +}
>>>>>> +
>>>>>> +#endif /* __LINUX_PRUSS_H */
>>>>>> -- 
>>>>>> 2.17.1
>>>>>>

Re: [PATCH v5 2/6] remoteproc: pru: Add APIs to get and put the PRU cores

[Romain Naour]

Hi Danish, All,

Le 07/09/2022 à 11:24, Md Danish Anwar a écrit :
> Hi Mathieu,
> 
> On 07/09/22 00:58, Mathieu Poirier wrote:
>> On Fri, Sep 02, 2022 at 03:09:04PM +0530, Md Danish Anwar wrote:
>>> Hi Mathieu,
>>>
>>>
>>> On 15/07/22 11:52, Kishon Vijay Abraham I wrote:
>>>> +Danish
>>>>
>>>> Hi Mathieu,
>>>>
>>>> On 14/07/22 22:50, Mathieu Poirier wrote:
>>>>> On Tue, Jul 12, 2022 at 11:59:49AM -0600, Mathieu Poirier wrote:
>>>>>> Hi Puranjay,
>>>>
>>>> Removed Puranjay (as he is no longer with TI) and adding Danish.
>>>>
>>>> Regards,
>>>> Kishon
>>>>>>
>>>>>> On Tue, Jun 07, 2022 at 10:26:46AM +0530, Puranjay Mohan wrote:
>>>>>>> From: Tero Kristo <t-kristo@ti.com>
>>>>>>>
>>>>>>> Add two new APIs, pru_rproc_get() and pru_rproc_put(), to the PRU
>>>>>>> driver to allow client drivers to acquire and release the remoteproc
>>>>>>> device associated with a PRU core. The PRU cores are treated as
>>>>>>> resources with only one client owning it at a time.
>>>>>>>
>>>>>>> The pru_rproc_get() function returns the rproc handle corresponding
>>>>>>> to a PRU core identified by the device tree "ti,prus" property under
>>>>>>> the client node. The pru_rproc_put() is the complementary function
>>>>>>> to pru_rproc_get().
>>>>>>>
>>>>>>> Co-developed-by: Suman Anna <s-anna@ti.com>
>>>>>>> Signed-off-by: Suman Anna <s-anna@ti.com>
>>>>>>> Signed-off-by: Tero Kristo <t-kristo@ti.com>
>>>>>>> Co-developed-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
>>>>>>> Signed-off-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
>>>>>>> Co-developed-by: Puranjay Mohan <p-mohan@ti.com>
>>>>>>> Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
>>>>>>> ---
>>>>>>>   drivers/remoteproc/pru_rproc.c | 138 +++++++++++++++++++++++++++++++--
>>>>>>>   include/linux/pruss.h          |  56 +++++++++++++
>>>>>>>   2 files changed, 189 insertions(+), 5 deletions(-)
>>>>>>>   create mode 100644 include/linux/pruss.h
>>>>>>>
>>>>>>> diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
>>>>>>> index 1777a01fa84e..7a35b400287a 100644
>>>>>>> --- a/drivers/remoteproc/pru_rproc.c
>>>>>>> +++ b/drivers/remoteproc/pru_rproc.c
>>>>>>> @@ -2,12 +2,13 @@
>>>>>>>   /*
>>>>>>>    * PRU-ICSS remoteproc driver for various TI SoCs
>>>>>>>    *
>>>>>>> - * Copyright (C) 2014-2020 Texas Instruments Incorporated - https://www.ti.com/
>>>>>>> + * Copyright (C) 2014-2022 Texas Instruments Incorporated - https://www.ti.com/
>>>>>>>    *
>>>>>>>    * Author(s):
>>>>>>>    *	Suman Anna <s-anna@ti.com>
>>>>>>>    *	Andrew F. Davis <afd@ti.com>
>>>>>>>    *	Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org> for Texas Instruments
>>>>>>> + *	Puranjay Mohan <p-mohan@ti.com>
>>>>>>>    */
>>>>>>>   #include <linux/bitops.h>
>>>>>>> @@ -16,6 +17,7 @@
>>>>>>>   #include <linux/module.h>
>>>>>>>   #include <linux/of_device.h>
>>>>>>>   #include <linux/of_irq.h>
>>>>>>> +#include <linux/pruss.h>
>>>>>>>   #include <linux/pruss_driver.h>
>>>>>>>   #include <linux/remoteproc.h>
>>>>>>> @@ -111,6 +113,8 @@ struct pru_private_data {
>>>>>>>    * @rproc: remoteproc pointer for this PRU core
>>>>>>>    * @data: PRU core specific data
>>>>>>>    * @mem_regions: data for each of the PRU memory regions
>>>>>>> + * @client_np: client device node
>>>>>>> + * @lock: mutex to protect client usage
>>>>>>>    * @fw_name: name of firmware image used during loading
>>>>>>>    * @mapped_irq: virtual interrupt numbers of created fw specific mapping
>>>>>>>    * @pru_interrupt_map: pointer to interrupt mapping description (firmware)
>>>>>>> @@ -126,6 +130,8 @@ struct pru_rproc {
>>>>>>>   	struct rproc *rproc;
>>>>>>>   	const struct pru_private_data *data;
>>>>>>>   	struct pruss_mem_region mem_regions[PRU_IOMEM_MAX];
>>>>>>> +	struct device_node *client_np;
>>>>>>> +	struct mutex lock; /* client access lock */
>>>>>>>   	const char *fw_name;
>>>>>>>   	unsigned int *mapped_irq;
>>>>>>>   	struct pru_irq_rsc *pru_interrupt_map;
>>>>>>> @@ -146,6 +152,125 @@ void pru_control_write_reg(struct pru_rproc *pru, unsigned int reg, u32 val)
>>>>>>>   	writel_relaxed(val, pru->mem_regions[PRU_IOMEM_CTRL].va + reg);
>>>>>>>   }
>>>>>>> +static struct rproc *__pru_rproc_get(struct device_node *np, int index)
>>>>>>> +{
>>>>>>> +	struct device_node *rproc_np = NULL;
>>>>>>> +	struct platform_device *pdev;
>>>>>>> +	struct rproc *rproc;
>>>>>>> +
>>>>>>> +	rproc_np = of_parse_phandle(np, "ti,prus", index);
>>>>>>> +	if (!rproc_np || !of_device_is_available(rproc_np))
>>>>>>> +		return ERR_PTR(-ENODEV);
>>>>>>> +
>>>>>>> +	pdev = of_find_device_by_node(rproc_np);
>>>>>>> +	of_node_put(rproc_np);
>>>>>>> +
>>>>>>> +	if (!pdev || !(&pdev->dev) || !((&pdev->dev)->driver))
>>>>>>> +		/* probably PRU not yet probed */
>>>>>>> +		return ERR_PTR(-EPROBE_DEFER);
>>>>>>> +
>>>>>>> +	/* make sure it is PRU rproc */
>>>>>>> +	if (!is_pru_rproc(&pdev->dev)) {
>>>>>>> +		put_device(&pdev->dev);
>>>>>>> +		return ERR_PTR(-ENODEV);
>>>>>>> +	}
>>>>>>> +
>>>>>>> +	rproc = platform_get_drvdata(pdev);
>>>>>>> +	put_device(&pdev->dev);
>>>>>>> +	if (!rproc)
>>>>>>> +		return ERR_PTR(-EPROBE_DEFER);
>>>>>>> +
>>>>>>> +	get_device(&rproc->dev);
>>>>>>> +
>>>>>>> +	return rproc;
>>>>>>> +}
>>>>>>> +
>>>>>>> +/**
>>>>>>> + * pru_rproc_get() - get the PRU rproc instance from a device node
>>>>>>> + * @np: the user/client device node
>>>>>>> + * @index: index to use for the ti,prus property
>>>>>>> + * @pru_id: optional pointer to return the PRU remoteproc processor id
>>>>>>> + *
>>>>>>> + * This function looks through a client device node's "ti,prus" property at
>>>>>>> + * index @index and returns the rproc handle for a valid PRU remote processor if
>>>>>>> + * found. The function allows only one user to own the PRU rproc resource at a
>>>>>>> + * time. Caller must call pru_rproc_put() when done with using the rproc, not
>>>>>>> + * required if the function returns a failure.
>>>>>>> + *
>>>>>>> + * When optional @pru_id pointer is passed the PRU remoteproc processor id is
>>>>>>> + * returned.
>>>>>>> + *
>>>>>>> + * Return: rproc handle on success, and an ERR_PTR on failure using one
>>>>>>> + * of the following error values
>>>>>>> + *    -ENODEV if device is not found
>>>>>>> + *    -EBUSY if PRU is already acquired by anyone
>>>>>>> + *    -EPROBE_DEFER is PRU device is not probed yet
>>>>>>> + */
>>>>>>> +struct rproc *pru_rproc_get(struct device_node *np, int index,
>>>>>>> +			    enum pruss_pru_id *pru_id)
>>>>>>> +{
>>>>>>> +	struct rproc *rproc;
>>>>>>> +	struct pru_rproc *pru;
>>>>>>> +	struct device *dev;
>>>>>>> +
>>>>>>> +	try_module_get(THIS_MODULE);
>>>>>>
>>>>>> There should be a module_put() in pru_rproc_put()...
>>>>>
>>>>> ... and in the error path of this function.
>>>>>
>>>>>>
>>>>>> More comments to come tomorrow.  I'm especially worried about this API racing
>>>>>> with a remote processor being removed or detached.
>>>>>>
>>>>>
>>>>> Looking at what is done in wkup_m3_ipc_probe(), it should be possible to call
>>>>> rproc_get_by_handle() here and that would make sure the remote processor doesn't
>>>>> go away before the end of the function.
>>>>>
>>>>> More comments to come...
>>>
>>> It is possible to call rproc_get_by_handle() here instead of
>>> __pru_get_proc(), but that would not provide multiple functionality.
>>>
>>> The API rproc_get_by_handle() returns rproc handle on success, and NULL on
>>> failure where as __pru_get_proc() returns ERR_PTR on failure which provides
>>> multiple functionality and opportunity for us to distinguish between
>>> multiple errors.
>>>
>>> So we have these three options.
>>>
>>> 1. If we're using the API rproc_get_by_handle() and we want the multiple
>>> ERR_PTR on failure then we will need to change the API rproc_get_by_handle()
>>> and also all the functions that uses rproc_get_by_handle().
>>>
>>
>> Not optimal.
>>
>>> 2. Keep the API rproc_get_by_handle() as it is. That will restrict us from
>>> using multiple ERR_PTR on different kinds of error.
>>>
>>
>> Not optimal.
>>
>>> 3. Instead of using rproc_get_by_handle(), keep using __pru_get_proc(). This
>>> will make sure we have the proper ERR_PTR to retrun for different kinds of
>>> errors.
>>>
>>
>> Unacceptable for the reason I already stated.  
>>
>>> Please let me know which option to continue with.
>>
>> I suggest building a wrapper that does everything you want around rproc_get_by_phandle().
>>
> 
> Sure, I'll do that. I'll share this change in v6 of this patch series.

I'm able to test the TI prueth driver from the ti-linux-5.10.y tree [1] on a
AM5749 cpu (custom board). But I need a more recent kernel (at least 5.15) to
support other devices recently added to the Linux kernel (wifi6 module and an
ethernet switch). Also it would be nice if this driver is finally merged in the
Linux kernel.

Maybe I can help to test this series but I noticed it only provide the driver
for TI AM654x cpus [2]. Can you also provide patches for basic EMAC support with
the TI AM574x too? (I don't need advanced features like frame timestamping, HSR
etc).

Also, what about patches present in the ti-linux-kernel tree and not included
this this series? Especially patches that modify the kernel network stack [3]
(net/rpmsg: add support for new rpmsg sockets). Is this new socket protocol
really needed?

Notice the patch adding the rpmsg sockets [3] already conflict with the upstream
kernel since the AF_MCTP definition now use the value temporarly used by
AF_RPMSG [4].

Can you send an updated version of the complete series?

Thanks!

[1] https://git.ti.com/cgit/ti-linux-kernel/ti-linux-kernel/log/?h=ti-linux-5.10.y

[2] https://lore.kernel.org/linux-remoteproc/20220406094358.7895-1-p-mohan@ti.com/

[3]
https://git.ti.com/cgit/ti-linux-kernel/ti-linux-kernel/commit/?h=ti-linux-5.10.y&id=f4b978a978c38149f712ddd137f12ed5fb914161

[4]
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=bc49d8169aa72295104f1558830c568efb946315

Best regards,
Romain


> 
>>>
>>> Thanks,
>>> Danish
>>>
>>>
>>>>>
>>>>>
>>>>>> Thanks,
>>>>>> Mathieu
>>>>>>
>>>>>>> +
>>>>>>> +	rproc = __pru_rproc_get(np, index);
>>>>>>> +	if (IS_ERR(rproc))
>>>>>>> +		return rproc;
>>>>>>> +
>>>>>>> +	pru = rproc->priv;
>>>>>>> +	dev = &rproc->dev;
>>>>>>> +
>>>>>>> +	mutex_lock(&pru->lock);
>>>>>>> +
>>>>>>> +	if (pru->client_np) {
>>>>>>> +		mutex_unlock(&pru->lock);
>>>>>>> +		put_device(dev);
>>>>>>> +		return ERR_PTR(-EBUSY);
>>>>>>> +	}
>>>>>>> +
>>>>>>> +	pru->client_np = np;
>>>>>>> +
>>>>>>> +	mutex_unlock(&pru->lock);
>>>>>>> +
>>>>>>> +	if (pru_id)
>>>>>>> +		*pru_id = pru->id;
>>>>>>> +
>>>>>>> +	return rproc;
>>>>>>> +}
>>>>>>> +EXPORT_SYMBOL_GPL(pru_rproc_get);
>>>>>>> +
>>>>>>> +/**
>>>>>>> + * pru_rproc_put() - release the PRU rproc resource
>>>>>>> + * @rproc: the rproc resource to release
>>>>>>> + *
>>>>>>> + * Releases the PRU rproc resource and makes it available to other
>>>>>>> + * users.
>>>>>>> + */
>>>>>>> +void pru_rproc_put(struct rproc *rproc)
>>>>>>> +{
>>>>>>> +	struct pru_rproc *pru;
>>>>>>> +
>>>>>>> +	if (IS_ERR_OR_NULL(rproc) || !is_pru_rproc(rproc->dev.parent))
>>>>>>> +		return;
>>>>>>> +
>>>>>>> +	pru = rproc->priv;
>>>>>>> +
>>>>>>> +	mutex_lock(&pru->lock);
>>>>>>> +
>>>>>>> +	if (!pru->client_np) {
>>>>>>> +		mutex_unlock(&pru->lock);
>>>>>>> +		return;
>>>>>>> +	}
>>>>>>> +
>>>>>>> +	pru->client_np = NULL;
>>>>>>> +	mutex_unlock(&pru->lock);
>>>>>>> +
>>>>>>> +	put_device(&rproc->dev);
>>>>>>> +}
>>>>>>> +EXPORT_SYMBOL_GPL(pru_rproc_put);
>>>>>>> +
>>>>>>>   static inline u32 pru_debug_read_reg(struct pru_rproc *pru, unsigned int reg)
>>>>>>>   {
>>>>>>>   	return readl_relaxed(pru->mem_regions[PRU_IOMEM_DEBUG].va + reg);
>>>>>>> @@ -438,7 +563,7 @@ static void *pru_d_da_to_va(struct pru_rproc *pru, u32 da, size_t len)
>>>>>>>   	dram0 = pruss->mem_regions[PRUSS_MEM_DRAM0];
>>>>>>>   	dram1 = pruss->mem_regions[PRUSS_MEM_DRAM1];
>>>>>>>   	/* PRU1 has its local RAM addresses reversed */
>>>>>>> -	if (pru->id == 1)
>>>>>>> +	if (pru->id == PRUSS_PRU1)
>>>>>>>   		swap(dram0, dram1);
>>>>>>>   	shrd_ram = pruss->mem_regions[PRUSS_MEM_SHRD_RAM2];
>>>>>>> @@ -747,14 +872,14 @@ static int pru_rproc_set_id(struct pru_rproc *pru)
>>>>>>>   	case RTU0_IRAM_ADDR_MASK:
>>>>>>>   		fallthrough;
>>>>>>>   	case PRU0_IRAM_ADDR_MASK:
>>>>>>> -		pru->id = 0;
>>>>>>> +		pru->id = PRUSS_PRU0;
>>>>>>>   		break;
>>>>>>>   	case TX_PRU1_IRAM_ADDR_MASK:
>>>>>>>   		fallthrough;
>>>>>>>   	case RTU1_IRAM_ADDR_MASK:
>>>>>>>   		fallthrough;
>>>>>>>   	case PRU1_IRAM_ADDR_MASK:
>>>>>>> -		pru->id = 1;
>>>>>>> +		pru->id = PRUSS_PRU1;
>>>>>>>   		break;
>>>>>>>   	default:
>>>>>>>   		ret = -EINVAL;
>>>>>>> @@ -816,6 +941,8 @@ static int pru_rproc_probe(struct platform_device *pdev)
>>>>>>>   	pru->pruss = platform_get_drvdata(ppdev);
>>>>>>>   	pru->rproc = rproc;
>>>>>>>   	pru->fw_name = fw_name;
>>>>>>> +	pru->client_np = NULL;
>>>>>>> +	mutex_init(&pru->lock);
>>>>>>>   	for (i = 0; i < ARRAY_SIZE(mem_names); i++) {
>>>>>>>   		res = platform_get_resource_byname(pdev, IORESOURCE_MEM,
>>>>>>> @@ -903,7 +1030,7 @@ MODULE_DEVICE_TABLE(of, pru_rproc_match);
>>>>>>>   static struct platform_driver pru_rproc_driver = {
>>>>>>>   	.driver = {
>>>>>>> -		.name   = "pru-rproc",
>>>>>>> +		.name   = PRU_RPROC_DRVNAME,
>>>>>>>   		.of_match_table = pru_rproc_match,
>>>>>>>   		.suppress_bind_attrs = true,
>>>>>>>   	},
>>>>>>> @@ -915,5 +1042,6 @@ module_platform_driver(pru_rproc_driver);
>>>>>>>   MODULE_AUTHOR("Suman Anna <s-anna@ti.com>");
>>>>>>>   MODULE_AUTHOR("Andrew F. Davis <afd@ti.com>");
>>>>>>>   MODULE_AUTHOR("Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>");
>>>>>>> +MODULE_AUTHOR("Puranjay Mohan <p-mohan@ti.com>");
>>>>>>>   MODULE_DESCRIPTION("PRU-ICSS Remote Processor Driver");
>>>>>>>   MODULE_LICENSE("GPL v2");
>>>>>>> diff --git a/include/linux/pruss.h b/include/linux/pruss.h
>>>>>>> new file mode 100644
>>>>>>> index 000000000000..fdc719b43db0
>>>>>>> --- /dev/null
>>>>>>> +++ b/include/linux/pruss.h
>>>>>>> @@ -0,0 +1,56 @@
>>>>>>> +/* SPDX-License-Identifier: GPL-2.0-only */
>>>>>>> +/**
>>>>>>> + * PRU-ICSS Subsystem user interfaces
>>>>>>> + *
>>>>>>> + * Copyright (C) 2015-2022 Texas Instruments Incorporated - http://www.ti.com
>>>>>>> + *	Suman Anna <s-anna@ti.com>
>>>>>>> + */
>>>>>>> +
>>>>>>> +#ifndef __LINUX_PRUSS_H
>>>>>>> +#define __LINUX_PRUSS_H
>>>>>>> +
>>>>>>> +#include <linux/device.h>
>>>>>>> +#include <linux/types.h>
>>>>>>> +
>>>>>>> +#define PRU_RPROC_DRVNAME "pru-rproc"
>>>>>>> +
>>>>>>> +/*
>>>>>>> + * enum pruss_pru_id - PRU core identifiers
>>>>>>> + */
>>>>>>> +enum pruss_pru_id {
>>>>>>> +	PRUSS_PRU0 = 0,
>>>>>>> +	PRUSS_PRU1,
>>>>>>> +	PRUSS_NUM_PRUS,
>>>>>>> +};
>>>>>>> +
>>>>>>> +struct device_node;
>>>>>>> +
>>>>>>> +#if IS_ENABLED(CONFIG_PRU_REMOTEPROC)
>>>>>>> +
>>>>>>> +struct rproc *pru_rproc_get(struct device_node *np, int index,
>>>>>>> +			    enum pruss_pru_id *pru_id);
>>>>>>> +void pru_rproc_put(struct rproc *rproc);
>>>>>>> +
>>>>>>> +#else
>>>>>>> +
>>>>>>> +static inline struct rproc *
>>>>>>> +pru_rproc_get(struct device_node *np, int index, enum pruss_pru_id *pru_id)
>>>>>>> +{
>>>>>>> +	return ERR_PTR(-EOPNOTSUPP);
>>>>>>> +}
>>>>>>> +
>>>>>>> +static inline void pru_rproc_put(struct rproc *rproc) { }
>>>>>>> +
>>>>>>> +#endif /* CONFIG_PRU_REMOTEPROC */
>>>>>>> +
>>>>>>> +static inline bool is_pru_rproc(struct device *dev)
>>>>>>> +{
>>>>>>> +	const char *drv_name = dev_driver_string(dev);
>>>>>>> +
>>>>>>> +	if (strncmp(drv_name, PRU_RPROC_DRVNAME, sizeof(PRU_RPROC_DRVNAME)))
>>>>>>> +		return false;
>>>>>>> +
>>>>>>> +	return true;
>>>>>>> +}
>>>>>>> +
>>>>>>> +#endif /* __LINUX_PRUSS_H */
>>>>>>> -- 
>>>>>>> 2.17.1
>>>>>>>
> 

Re: [PATCH v5 2/6] remoteproc: pru: Add APIs to get and put the PRU cores

[Md Danish Anwar]

Hi Mathieu,

On 12/09/22 18:54, Romain Naour wrote:
> Hi Danish, All,
> 
> Le 07/09/2022 à 11:24, Md Danish Anwar a écrit :
>> Hi Mathieu,
>>
>> On 07/09/22 00:58, Mathieu Poirier wrote:
>>> On Fri, Sep 02, 2022 at 03:09:04PM +0530, Md Danish Anwar wrote:
>>>> Hi Mathieu,
>>>>
>>>>
>>>> On 15/07/22 11:52, Kishon Vijay Abraham I wrote:
>>>>> +Danish
>>>>>
>>>>> Hi Mathieu,
>>>>>
>>>>> On 14/07/22 22:50, Mathieu Poirier wrote:
>>>>>> On Tue, Jul 12, 2022 at 11:59:49AM -0600, Mathieu Poirier wrote:
>>>>>>> Hi Puranjay,
>>>>>
>>>>> Removed Puranjay (as he is no longer with TI) and adding Danish.
>>>>>
>>>>> Regards,
>>>>> Kishon
>>>>>>>
>>>>>>> On Tue, Jun 07, 2022 at 10:26:46AM +0530, Puranjay Mohan wrote:
>>>>>>>> From: Tero Kristo <t-kristo@ti.com>
>>>>>>>>
>>>>>>>> Add two new APIs, pru_rproc_get() and pru_rproc_put(), to the PRU
>>>>>>>> driver to allow client drivers to acquire and release the remoteproc
>>>>>>>> device associated with a PRU core. The PRU cores are treated as
>>>>>>>> resources with only one client owning it at a time.
>>>>>>>>
>>>>>>>> The pru_rproc_get() function returns the rproc handle corresponding
>>>>>>>> to a PRU core identified by the device tree "ti,prus" property under
>>>>>>>> the client node. The pru_rproc_put() is the complementary function
>>>>>>>> to pru_rproc_get().
>>>>>>>>
>>>>>>>> Co-developed-by: Suman Anna <s-anna@ti.com>
>>>>>>>> Signed-off-by: Suman Anna <s-anna@ti.com>
>>>>>>>> Signed-off-by: Tero Kristo <t-kristo@ti.com>
>>>>>>>> Co-developed-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
>>>>>>>> Signed-off-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
>>>>>>>> Co-developed-by: Puranjay Mohan <p-mohan@ti.com>
>>>>>>>> Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
>>>>>>>> ---
>>>>>>>>   drivers/remoteproc/pru_rproc.c | 138 +++++++++++++++++++++++++++++++--
>>>>>>>>   include/linux/pruss.h          |  56 +++++++++++++
>>>>>>>>   2 files changed, 189 insertions(+), 5 deletions(-)
>>>>>>>>   create mode 100644 include/linux/pruss.h
>>>>>>>>
>>>>>>>> diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
>>>>>>>> index 1777a01fa84e..7a35b400287a 100644
>>>>>>>> --- a/drivers/remoteproc/pru_rproc.c
>>>>>>>> +++ b/drivers/remoteproc/pru_rproc.c
>>>>>>>> @@ -2,12 +2,13 @@
>>>>>>>>   /*
>>>>>>>>    * PRU-ICSS remoteproc driver for various TI SoCs
>>>>>>>>    *
>>>>>>>> - * Copyright (C) 2014-2020 Texas Instruments Incorporated - https://www.ti.com/
>>>>>>>> + * Copyright (C) 2014-2022 Texas Instruments Incorporated - https://www.ti.com/
>>>>>>>>    *
>>>>>>>>    * Author(s):
>>>>>>>>    *	Suman Anna <s-anna@ti.com>
>>>>>>>>    *	Andrew F. Davis <afd@ti.com>
>>>>>>>>    *	Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org> for Texas Instruments
>>>>>>>> + *	Puranjay Mohan <p-mohan@ti.com>
>>>>>>>>    */
>>>>>>>>   #include <linux/bitops.h>
>>>>>>>> @@ -16,6 +17,7 @@
>>>>>>>>   #include <linux/module.h>
>>>>>>>>   #include <linux/of_device.h>
>>>>>>>>   #include <linux/of_irq.h>
>>>>>>>> +#include <linux/pruss.h>
>>>>>>>>   #include <linux/pruss_driver.h>
>>>>>>>>   #include <linux/remoteproc.h>
>>>>>>>> @@ -111,6 +113,8 @@ struct pru_private_data {
>>>>>>>>    * @rproc: remoteproc pointer for this PRU core
>>>>>>>>    * @data: PRU core specific data
>>>>>>>>    * @mem_regions: data for each of the PRU memory regions
>>>>>>>> + * @client_np: client device node
>>>>>>>> + * @lock: mutex to protect client usage
>>>>>>>>    * @fw_name: name of firmware image used during loading
>>>>>>>>    * @mapped_irq: virtual interrupt numbers of created fw specific mapping
>>>>>>>>    * @pru_interrupt_map: pointer to interrupt mapping description (firmware)
>>>>>>>> @@ -126,6 +130,8 @@ struct pru_rproc {
>>>>>>>>   	struct rproc *rproc;
>>>>>>>>   	const struct pru_private_data *data;
>>>>>>>>   	struct pruss_mem_region mem_regions[PRU_IOMEM_MAX];
>>>>>>>> +	struct device_node *client_np;
>>>>>>>> +	struct mutex lock; /* client access lock */
>>>>>>>>   	const char *fw_name;
>>>>>>>>   	unsigned int *mapped_irq;
>>>>>>>>   	struct pru_irq_rsc *pru_interrupt_map;
>>>>>>>> @@ -146,6 +152,125 @@ void pru_control_write_reg(struct pru_rproc *pru, unsigned int reg, u32 val)
>>>>>>>>   	writel_relaxed(val, pru->mem_regions[PRU_IOMEM_CTRL].va + reg);
>>>>>>>>   }
>>>>>>>> +static struct rproc *__pru_rproc_get(struct device_node *np, int index)
>>>>>>>> +{
>>>>>>>> +	struct device_node *rproc_np = NULL;
>>>>>>>> +	struct platform_device *pdev;
>>>>>>>> +	struct rproc *rproc;
>>>>>>>> +
>>>>>>>> +	rproc_np = of_parse_phandle(np, "ti,prus", index);
>>>>>>>> +	if (!rproc_np || !of_device_is_available(rproc_np))
>>>>>>>> +		return ERR_PTR(-ENODEV);
>>>>>>>> +
>>>>>>>> +	pdev = of_find_device_by_node(rproc_np);
>>>>>>>> +	of_node_put(rproc_np);
>>>>>>>> +
>>>>>>>> +	if (!pdev || !(&pdev->dev) || !((&pdev->dev)->driver))
>>>>>>>> +		/* probably PRU not yet probed */
>>>>>>>> +		return ERR_PTR(-EPROBE_DEFER);
>>>>>>>> +
>>>>>>>> +	/* make sure it is PRU rproc */
>>>>>>>> +	if (!is_pru_rproc(&pdev->dev)) {
>>>>>>>> +		put_device(&pdev->dev);
>>>>>>>> +		return ERR_PTR(-ENODEV);
>>>>>>>> +	}
>>>>>>>> +
>>>>>>>> +	rproc = platform_get_drvdata(pdev);
>>>>>>>> +	put_device(&pdev->dev);
>>>>>>>> +	if (!rproc)
>>>>>>>> +		return ERR_PTR(-EPROBE_DEFER);
>>>>>>>> +
>>>>>>>> +	get_device(&rproc->dev);
>>>>>>>> +
>>>>>>>> +	return rproc;
>>>>>>>> +}
>>>>>>>> +
>>>>>>>> +/**
>>>>>>>> + * pru_rproc_get() - get the PRU rproc instance from a device node
>>>>>>>> + * @np: the user/client device node
>>>>>>>> + * @index: index to use for the ti,prus property
>>>>>>>> + * @pru_id: optional pointer to return the PRU remoteproc processor id
>>>>>>>> + *
>>>>>>>> + * This function looks through a client device node's "ti,prus" property at
>>>>>>>> + * index @index and returns the rproc handle for a valid PRU remote processor if
>>>>>>>> + * found. The function allows only one user to own the PRU rproc resource at a
>>>>>>>> + * time. Caller must call pru_rproc_put() when done with using the rproc, not
>>>>>>>> + * required if the function returns a failure.
>>>>>>>> + *
>>>>>>>> + * When optional @pru_id pointer is passed the PRU remoteproc processor id is
>>>>>>>> + * returned.
>>>>>>>> + *
>>>>>>>> + * Return: rproc handle on success, and an ERR_PTR on failure using one
>>>>>>>> + * of the following error values
>>>>>>>> + *    -ENODEV if device is not found
>>>>>>>> + *    -EBUSY if PRU is already acquired by anyone
>>>>>>>> + *    -EPROBE_DEFER is PRU device is not probed yet
>>>>>>>> + */
>>>>>>>> +struct rproc *pru_rproc_get(struct device_node *np, int index,
>>>>>>>> +			    enum pruss_pru_id *pru_id)
>>>>>>>> +{
>>>>>>>> +	struct rproc *rproc;
>>>>>>>> +	struct pru_rproc *pru;
>>>>>>>> +	struct device *dev;
>>>>>>>> +
>>>>>>>> +	try_module_get(THIS_MODULE);
>>>>>>>
>>>>>>> There should be a module_put() in pru_rproc_put()...
>>>>>>
>>>>>> ... and in the error path of this function.
>>>>>>
>>>>>>>
>>>>>>> More comments to come tomorrow.  I'm especially worried about this API racing
>>>>>>> with a remote processor being removed or detached.
>>>>>>>
>>>>>>
>>>>>> Looking at what is done in wkup_m3_ipc_probe(), it should be possible to call
>>>>>> rproc_get_by_handle() here and that would make sure the remote processor doesn't
>>>>>> go away before the end of the function.
>>>>>>
>>>>>> More comments to come...
>>>>
>>>> It is possible to call rproc_get_by_handle() here instead of
>>>> __pru_get_proc(), but that would not provide multiple functionality.
>>>>
>>>> The API rproc_get_by_handle() returns rproc handle on success, and NULL on
>>>> failure where as __pru_get_proc() returns ERR_PTR on failure which provides
>>>> multiple functionality and opportunity for us to distinguish between
>>>> multiple errors.
>>>>
>>>> So we have these three options.
>>>>
>>>> 1. If we're using the API rproc_get_by_handle() and we want the multiple
>>>> ERR_PTR on failure then we will need to change the API rproc_get_by_handle()
>>>> and also all the functions that uses rproc_get_by_handle().
>>>>
>>>
>>> Not optimal.
>>>
>>>> 2. Keep the API rproc_get_by_handle() as it is. That will restrict us from
>>>> using multiple ERR_PTR on different kinds of error.
>>>>
>>>
>>> Not optimal.
>>>
>>>> 3. Instead of using rproc_get_by_handle(), keep using __pru_get_proc(). This
>>>> will make sure we have the proper ERR_PTR to retrun for different kinds of
>>>> errors.
>>>>
>>>
>>> Unacceptable for the reason I already stated.  
>>>
>>>> Please let me know which option to continue with.
>>>
>>> I suggest building a wrapper that does everything you want around rproc_get_by_phandle().

We can introduce a new API __rproc_get_by_phandle() similar to the API
rproc_get_by_phandle(). The new API __rproc_get_by_phandle() will do all the
functionality of getting the rproc. On success it will return rproc and on
failure it will return the different ERR_PTR.
If rproc is not probed yet, it will return ERR_PTR(-EPROBE_DEFER).

This will make sure that we're getting different error codes for different
errors from pru_rproc_get().

The old API rproc_get_by_handle() will invoke the new API. On success the new
API will return rproc and the old API will also return rproc. On failure the
new API will return different error codes while the old API will preserve it's
nature and return NULL.

The API __rproc_get_by_handle will look like this.

struct rproc *__rproc_get_by_handle(phandle phandle)
{
	struct rproc *rproc = NULL, *r;
	struct device_node *np;

	np = of_find_node_by_phandle(phandle);
	if (!np)
		return NULL;

	rcu_read_lock();
	list_for_each_entry_rcu(r, &rproc_list, node) {
		if (r->dev.parent && r->dev.parent->of_node == np) {
			/*prevent underlying implementation from being removed */
			if (!try_module_get(r->dev.parent->driver->owner)) {
				dev_err(&r->dev, "can't get owner\n");
				break;
			}

			rproc = r;
			get_device(&rproc->dev);
			break;
		}
	}
	rcu_read_unlock();

	of_node_put(np);

	if(!rproc)
		return ERR_PTR(-EPROBE_DEFER);

	return rproc;
}

The API rproc_get_by_handle() will look like this.

struct rproc *rproc_get_by_phandle(phandle phandle)
{
	struct rproc *rproc = NULL;

	rproc = __rproc_get_by_handle(phandle);

	if(!rproc || IS_ERR(rproc))
		return NULL;

	return rproc;
}

This way in pru_rproc_get(), we'll get the rproc by phandle and we'll still
return different error codes depending upon failure cases. We'll also be able
to preserve the actual functionality of rproc_get_by_phandle() so that the
other APIs using rproc_get_by_phandle() won't get affected.

Please let me know if this looks good.

>>>
>>
>> Sure, I'll do that. I'll share this change in v6 of this patch series.
> 
> I'm able to test the TI prueth driver from the ti-linux-5.10.y tree [1] on a
> AM5749 cpu (custom board). But I need a more recent kernel (at least 5.15) to
> support other devices recently added to the Linux kernel (wifi6 module and an
> ethernet switch). Also it would be nice if this driver is finally merged in the
> Linux kernel.
> 
> Maybe I can help to test this series but I noticed it only provide the driver
> for TI AM654x cpus [2]. Can you also provide patches for basic EMAC support with
> the TI AM574x too? (I don't need advanced features like frame timestamping, HSR
> etc).
> 
> Also, what about patches present in the ti-linux-kernel tree and not included
> this this series? Especially patches that modify the kernel network stack [3]
> (net/rpmsg: add support for new rpmsg sockets). Is this new socket protocol
> really needed?
> 
> Notice the patch adding the rpmsg sockets [3] already conflict with the upstream
> kernel since the AF_MCTP definition now use the value temporarly used by
> AF_RPMSG [4].
> 
> Can you send an updated version of the complete series?
> 
> Thanks!
> 
> [1] https://git.ti.com/cgit/ti-linux-kernel/ti-linux-kernel/log/?h=ti-linux-5.10.y
> 
> [2] https://lore.kernel.org/linux-remoteproc/20220406094358.7895-1-p-mohan@ti.com/
> 
> [3]
> https://git.ti.com/cgit/ti-linux-kernel/ti-linux-kernel/commit/?h=ti-linux-5.10.y&id=f4b978a978c38149f712ddd137f12ed5fb914161
> 
> [4]
> https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=bc49d8169aa72295104f1558830c568efb946315
> 
> Best regards,
> Romain
> 
> 
>>
>>>>
>>>> Thanks,
>>>> Danish
>>>>
>>>>
>>>>>>
>>>>>>
>>>>>>> Thanks,
>>>>>>> Mathieu
>>>>>>>
>>>>>>>> +
>>>>>>>> +	rproc = __pru_rproc_get(np, index);
>>>>>>>> +	if (IS_ERR(rproc))
>>>>>>>> +		return rproc;
>>>>>>>> +
>>>>>>>> +	pru = rproc->priv;
>>>>>>>> +	dev = &rproc->dev;
>>>>>>>> +
>>>>>>>> +	mutex_lock(&pru->lock);
>>>>>>>> +
>>>>>>>> +	if (pru->client_np) {
>>>>>>>> +		mutex_unlock(&pru->lock);
>>>>>>>> +		put_device(dev);
>>>>>>>> +		return ERR_PTR(-EBUSY);
>>>>>>>> +	}
>>>>>>>> +
>>>>>>>> +	pru->client_np = np;
>>>>>>>> +
>>>>>>>> +	mutex_unlock(&pru->lock);
>>>>>>>> +
>>>>>>>> +	if (pru_id)
>>>>>>>> +		*pru_id = pru->id;
>>>>>>>> +
>>>>>>>> +	return rproc;
>>>>>>>> +}
>>>>>>>> +EXPORT_SYMBOL_GPL(pru_rproc_get);
>>>>>>>> +
>>>>>>>> +/**
>>>>>>>> + * pru_rproc_put() - release the PRU rproc resource
>>>>>>>> + * @rproc: the rproc resource to release
>>>>>>>> + *
>>>>>>>> + * Releases the PRU rproc resource and makes it available to other
>>>>>>>> + * users.
>>>>>>>> + */
>>>>>>>> +void pru_rproc_put(struct rproc *rproc)
>>>>>>>> +{
>>>>>>>> +	struct pru_rproc *pru;
>>>>>>>> +
>>>>>>>> +	if (IS_ERR_OR_NULL(rproc) || !is_pru_rproc(rproc->dev.parent))
>>>>>>>> +		return;
>>>>>>>> +
>>>>>>>> +	pru = rproc->priv;
>>>>>>>> +
>>>>>>>> +	mutex_lock(&pru->lock);
>>>>>>>> +
>>>>>>>> +	if (!pru->client_np) {
>>>>>>>> +		mutex_unlock(&pru->lock);
>>>>>>>> +		return;
>>>>>>>> +	}
>>>>>>>> +
>>>>>>>> +	pru->client_np = NULL;
>>>>>>>> +	mutex_unlock(&pru->lock);
>>>>>>>> +
>>>>>>>> +	put_device(&rproc->dev);
>>>>>>>> +}
>>>>>>>> +EXPORT_SYMBOL_GPL(pru_rproc_put);
>>>>>>>> +
>>>>>>>>   static inline u32 pru_debug_read_reg(struct pru_rproc *pru, unsigned int reg)
>>>>>>>>   {
>>>>>>>>   	return readl_relaxed(pru->mem_regions[PRU_IOMEM_DEBUG].va + reg);
>>>>>>>> @@ -438,7 +563,7 @@ static void *pru_d_da_to_va(struct pru_rproc *pru, u32 da, size_t len)
>>>>>>>>   	dram0 = pruss->mem_regions[PRUSS_MEM_DRAM0];
>>>>>>>>   	dram1 = pruss->mem_regions[PRUSS_MEM_DRAM1];
>>>>>>>>   	/* PRU1 has its local RAM addresses reversed */
>>>>>>>> -	if (pru->id == 1)
>>>>>>>> +	if (pru->id == PRUSS_PRU1)
>>>>>>>>   		swap(dram0, dram1);
>>>>>>>>   	shrd_ram = pruss->mem_regions[PRUSS_MEM_SHRD_RAM2];
>>>>>>>> @@ -747,14 +872,14 @@ static int pru_rproc_set_id(struct pru_rproc *pru)
>>>>>>>>   	case RTU0_IRAM_ADDR_MASK:
>>>>>>>>   		fallthrough;
>>>>>>>>   	case PRU0_IRAM_ADDR_MASK:
>>>>>>>> -		pru->id = 0;
>>>>>>>> +		pru->id = PRUSS_PRU0;
>>>>>>>>   		break;
>>>>>>>>   	case TX_PRU1_IRAM_ADDR_MASK:
>>>>>>>>   		fallthrough;
>>>>>>>>   	case RTU1_IRAM_ADDR_MASK:
>>>>>>>>   		fallthrough;
>>>>>>>>   	case PRU1_IRAM_ADDR_MASK:
>>>>>>>> -		pru->id = 1;
>>>>>>>> +		pru->id = PRUSS_PRU1;
>>>>>>>>   		break;
>>>>>>>>   	default:
>>>>>>>>   		ret = -EINVAL;
>>>>>>>> @@ -816,6 +941,8 @@ static int pru_rproc_probe(struct platform_device *pdev)
>>>>>>>>   	pru->pruss = platform_get_drvdata(ppdev);
>>>>>>>>   	pru->rproc = rproc;
>>>>>>>>   	pru->fw_name = fw_name;
>>>>>>>> +	pru->client_np = NULL;
>>>>>>>> +	mutex_init(&pru->lock);
>>>>>>>>   	for (i = 0; i < ARRAY_SIZE(mem_names); i++) {
>>>>>>>>   		res = platform_get_resource_byname(pdev, IORESOURCE_MEM,
>>>>>>>> @@ -903,7 +1030,7 @@ MODULE_DEVICE_TABLE(of, pru_rproc_match);
>>>>>>>>   static struct platform_driver pru_rproc_driver = {
>>>>>>>>   	.driver = {
>>>>>>>> -		.name   = "pru-rproc",
>>>>>>>> +		.name   = PRU_RPROC_DRVNAME,
>>>>>>>>   		.of_match_table = pru_rproc_match,
>>>>>>>>   		.suppress_bind_attrs = true,
>>>>>>>>   	},
>>>>>>>> @@ -915,5 +1042,6 @@ module_platform_driver(pru_rproc_driver);
>>>>>>>>   MODULE_AUTHOR("Suman Anna <s-anna@ti.com>");
>>>>>>>>   MODULE_AUTHOR("Andrew F. Davis <afd@ti.com>");
>>>>>>>>   MODULE_AUTHOR("Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>");
>>>>>>>> +MODULE_AUTHOR("Puranjay Mohan <p-mohan@ti.com>");
>>>>>>>>   MODULE_DESCRIPTION("PRU-ICSS Remote Processor Driver");
>>>>>>>>   MODULE_LICENSE("GPL v2");
>>>>>>>> diff --git a/include/linux/pruss.h b/include/linux/pruss.h
>>>>>>>> new file mode 100644
>>>>>>>> index 000000000000..fdc719b43db0
>>>>>>>> --- /dev/null
>>>>>>>> +++ b/include/linux/pruss.h
>>>>>>>> @@ -0,0 +1,56 @@
>>>>>>>> +/* SPDX-License-Identifier: GPL-2.0-only */
>>>>>>>> +/**
>>>>>>>> + * PRU-ICSS Subsystem user interfaces
>>>>>>>> + *
>>>>>>>> + * Copyright (C) 2015-2022 Texas Instruments Incorporated - http://www.ti.com
>>>>>>>> + *	Suman Anna <s-anna@ti.com>
>>>>>>>> + */
>>>>>>>> +
>>>>>>>> +#ifndef __LINUX_PRUSS_H
>>>>>>>> +#define __LINUX_PRUSS_H
>>>>>>>> +
>>>>>>>> +#include <linux/device.h>
>>>>>>>> +#include <linux/types.h>
>>>>>>>> +
>>>>>>>> +#define PRU_RPROC_DRVNAME "pru-rproc"
>>>>>>>> +
>>>>>>>> +/*
>>>>>>>> + * enum pruss_pru_id - PRU core identifiers
>>>>>>>> + */
>>>>>>>> +enum pruss_pru_id {
>>>>>>>> +	PRUSS_PRU0 = 0,
>>>>>>>> +	PRUSS_PRU1,
>>>>>>>> +	PRUSS_NUM_PRUS,
>>>>>>>> +};
>>>>>>>> +
>>>>>>>> +struct device_node;
>>>>>>>> +
>>>>>>>> +#if IS_ENABLED(CONFIG_PRU_REMOTEPROC)
>>>>>>>> +
>>>>>>>> +struct rproc *pru_rproc_get(struct device_node *np, int index,
>>>>>>>> +			    enum pruss_pru_id *pru_id);
>>>>>>>> +void pru_rproc_put(struct rproc *rproc);
>>>>>>>> +
>>>>>>>> +#else
>>>>>>>> +
>>>>>>>> +static inline struct rproc *
>>>>>>>> +pru_rproc_get(struct device_node *np, int index, enum pruss_pru_id *pru_id)
>>>>>>>> +{
>>>>>>>> +	return ERR_PTR(-EOPNOTSUPP);
>>>>>>>> +}
>>>>>>>> +
>>>>>>>> +static inline void pru_rproc_put(struct rproc *rproc) { }
>>>>>>>> +
>>>>>>>> +#endif /* CONFIG_PRU_REMOTEPROC */
>>>>>>>> +
>>>>>>>> +static inline bool is_pru_rproc(struct device *dev)
>>>>>>>> +{
>>>>>>>> +	const char *drv_name = dev_driver_string(dev);
>>>>>>>> +
>>>>>>>> +	if (strncmp(drv_name, PRU_RPROC_DRVNAME, sizeof(PRU_RPROC_DRVNAME)))
>>>>>>>> +		return false;
>>>>>>>> +
>>>>>>>> +	return true;
>>>>>>>> +}
>>>>>>>> +
>>>>>>>> +#endif /* __LINUX_PRUSS_H */
>>>>>>>> -- 
>>>>>>>> 2.17.1
>>>>>>>>
>>
> 

Re: [PATCH v5 2/6] remoteproc: pru: Add APIs to get and put the PRU cores

[Mathieu Poirier]

On Tue, 13 Sept 2022 at 05:40, Md Danish Anwar <a0501179@ti.com> wrote:
>
> Hi Mathieu,
>
> On 12/09/22 18:54, Romain Naour wrote:
> > Hi Danish, All,
> >
> > Le 07/09/2022 à 11:24, Md Danish Anwar a écrit :
> >> Hi Mathieu,
> >>
> >> On 07/09/22 00:58, Mathieu Poirier wrote:
> >>> On Fri, Sep 02, 2022 at 03:09:04PM +0530, Md Danish Anwar wrote:
> >>>> Hi Mathieu,
> >>>>
> >>>>
> >>>> On 15/07/22 11:52, Kishon Vijay Abraham I wrote:
> >>>>> +Danish
> >>>>>
> >>>>> Hi Mathieu,
> >>>>>
> >>>>> On 14/07/22 22:50, Mathieu Poirier wrote:
> >>>>>> On Tue, Jul 12, 2022 at 11:59:49AM -0600, Mathieu Poirier wrote:
> >>>>>>> Hi Puranjay,
> >>>>>
> >>>>> Removed Puranjay (as he is no longer with TI) and adding Danish.
> >>>>>
> >>>>> Regards,
> >>>>> Kishon
> >>>>>>>
> >>>>>>> On Tue, Jun 07, 2022 at 10:26:46AM +0530, Puranjay Mohan wrote:
> >>>>>>>> From: Tero Kristo <t-kristo@ti.com>
> >>>>>>>>
> >>>>>>>> Add two new APIs, pru_rproc_get() and pru_rproc_put(), to the PRU
> >>>>>>>> driver to allow client drivers to acquire and release the remoteproc
> >>>>>>>> device associated with a PRU core. The PRU cores are treated as
> >>>>>>>> resources with only one client owning it at a time.
> >>>>>>>>
> >>>>>>>> The pru_rproc_get() function returns the rproc handle corresponding
> >>>>>>>> to a PRU core identified by the device tree "ti,prus" property under
> >>>>>>>> the client node. The pru_rproc_put() is the complementary function
> >>>>>>>> to pru_rproc_get().
> >>>>>>>>
> >>>>>>>> Co-developed-by: Suman Anna <s-anna@ti.com>
> >>>>>>>> Signed-off-by: Suman Anna <s-anna@ti.com>
> >>>>>>>> Signed-off-by: Tero Kristo <t-kristo@ti.com>
> >>>>>>>> Co-developed-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
> >>>>>>>> Signed-off-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
> >>>>>>>> Co-developed-by: Puranjay Mohan <p-mohan@ti.com>
> >>>>>>>> Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
> >>>>>>>> ---
> >>>>>>>>   drivers/remoteproc/pru_rproc.c | 138 +++++++++++++++++++++++++++++++--
> >>>>>>>>   include/linux/pruss.h          |  56 +++++++++++++
> >>>>>>>>   2 files changed, 189 insertions(+), 5 deletions(-)
> >>>>>>>>   create mode 100644 include/linux/pruss.h
> >>>>>>>>
> >>>>>>>> diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
> >>>>>>>> index 1777a01fa84e..7a35b400287a 100644
> >>>>>>>> --- a/drivers/remoteproc/pru_rproc.c
> >>>>>>>> +++ b/drivers/remoteproc/pru_rproc.c
> >>>>>>>> @@ -2,12 +2,13 @@
> >>>>>>>>   /*
> >>>>>>>>    * PRU-ICSS remoteproc driver for various TI SoCs
> >>>>>>>>    *
> >>>>>>>> - * Copyright (C) 2014-2020 Texas Instruments Incorporated - https://www.ti.com/
> >>>>>>>> + * Copyright (C) 2014-2022 Texas Instruments Incorporated - https://www.ti.com/
> >>>>>>>>    *
> >>>>>>>>    * Author(s):
> >>>>>>>>    *   Suman Anna <s-anna@ti.com>
> >>>>>>>>    *   Andrew F. Davis <afd@ti.com>
> >>>>>>>>    *   Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org> for Texas Instruments
> >>>>>>>> + *    Puranjay Mohan <p-mohan@ti.com>
> >>>>>>>>    */
> >>>>>>>>   #include <linux/bitops.h>
> >>>>>>>> @@ -16,6 +17,7 @@
> >>>>>>>>   #include <linux/module.h>
> >>>>>>>>   #include <linux/of_device.h>
> >>>>>>>>   #include <linux/of_irq.h>
> >>>>>>>> +#include <linux/pruss.h>
> >>>>>>>>   #include <linux/pruss_driver.h>
> >>>>>>>>   #include <linux/remoteproc.h>
> >>>>>>>> @@ -111,6 +113,8 @@ struct pru_private_data {
> >>>>>>>>    * @rproc: remoteproc pointer for this PRU core
> >>>>>>>>    * @data: PRU core specific data
> >>>>>>>>    * @mem_regions: data for each of the PRU memory regions
> >>>>>>>> + * @client_np: client device node
> >>>>>>>> + * @lock: mutex to protect client usage
> >>>>>>>>    * @fw_name: name of firmware image used during loading
> >>>>>>>>    * @mapped_irq: virtual interrupt numbers of created fw specific mapping
> >>>>>>>>    * @pru_interrupt_map: pointer to interrupt mapping description (firmware)
> >>>>>>>> @@ -126,6 +130,8 @@ struct pru_rproc {
> >>>>>>>>        struct rproc *rproc;
> >>>>>>>>        const struct pru_private_data *data;
> >>>>>>>>        struct pruss_mem_region mem_regions[PRU_IOMEM_MAX];
> >>>>>>>> +      struct device_node *client_np;
> >>>>>>>> +      struct mutex lock; /* client access lock */
> >>>>>>>>        const char *fw_name;
> >>>>>>>>        unsigned int *mapped_irq;
> >>>>>>>>        struct pru_irq_rsc *pru_interrupt_map;
> >>>>>>>> @@ -146,6 +152,125 @@ void pru_control_write_reg(struct pru_rproc *pru, unsigned int reg, u32 val)
> >>>>>>>>        writel_relaxed(val, pru->mem_regions[PRU_IOMEM_CTRL].va + reg);
> >>>>>>>>   }
> >>>>>>>> +static struct rproc *__pru_rproc_get(struct device_node *np, int index)
> >>>>>>>> +{
> >>>>>>>> +      struct device_node *rproc_np = NULL;
> >>>>>>>> +      struct platform_device *pdev;
> >>>>>>>> +      struct rproc *rproc;
> >>>>>>>> +
> >>>>>>>> +      rproc_np = of_parse_phandle(np, "ti,prus", index);
> >>>>>>>> +      if (!rproc_np || !of_device_is_available(rproc_np))
> >>>>>>>> +              return ERR_PTR(-ENODEV);
> >>>>>>>> +
> >>>>>>>> +      pdev = of_find_device_by_node(rproc_np);
> >>>>>>>> +      of_node_put(rproc_np);
> >>>>>>>> +
> >>>>>>>> +      if (!pdev || !(&pdev->dev) || !((&pdev->dev)->driver))
> >>>>>>>> +              /* probably PRU not yet probed */
> >>>>>>>> +              return ERR_PTR(-EPROBE_DEFER);
> >>>>>>>> +
> >>>>>>>> +      /* make sure it is PRU rproc */
> >>>>>>>> +      if (!is_pru_rproc(&pdev->dev)) {
> >>>>>>>> +              put_device(&pdev->dev);
> >>>>>>>> +              return ERR_PTR(-ENODEV);
> >>>>>>>> +      }
> >>>>>>>> +
> >>>>>>>> +      rproc = platform_get_drvdata(pdev);
> >>>>>>>> +      put_device(&pdev->dev);
> >>>>>>>> +      if (!rproc)
> >>>>>>>> +              return ERR_PTR(-EPROBE_DEFER);
> >>>>>>>> +
> >>>>>>>> +      get_device(&rproc->dev);
> >>>>>>>> +
> >>>>>>>> +      return rproc;
> >>>>>>>> +}
> >>>>>>>> +
> >>>>>>>> +/**
> >>>>>>>> + * pru_rproc_get() - get the PRU rproc instance from a device node
> >>>>>>>> + * @np: the user/client device node
> >>>>>>>> + * @index: index to use for the ti,prus property
> >>>>>>>> + * @pru_id: optional pointer to return the PRU remoteproc processor id
> >>>>>>>> + *
> >>>>>>>> + * This function looks through a client device node's "ti,prus" property at
> >>>>>>>> + * index @index and returns the rproc handle for a valid PRU remote processor if
> >>>>>>>> + * found. The function allows only one user to own the PRU rproc resource at a
> >>>>>>>> + * time. Caller must call pru_rproc_put() when done with using the rproc, not
> >>>>>>>> + * required if the function returns a failure.
> >>>>>>>> + *
> >>>>>>>> + * When optional @pru_id pointer is passed the PRU remoteproc processor id is
> >>>>>>>> + * returned.
> >>>>>>>> + *
> >>>>>>>> + * Return: rproc handle on success, and an ERR_PTR on failure using one
> >>>>>>>> + * of the following error values
> >>>>>>>> + *    -ENODEV if device is not found
> >>>>>>>> + *    -EBUSY if PRU is already acquired by anyone
> >>>>>>>> + *    -EPROBE_DEFER is PRU device is not probed yet
> >>>>>>>> + */
> >>>>>>>> +struct rproc *pru_rproc_get(struct device_node *np, int index,
> >>>>>>>> +                          enum pruss_pru_id *pru_id)
> >>>>>>>> +{
> >>>>>>>> +      struct rproc *rproc;
> >>>>>>>> +      struct pru_rproc *pru;
> >>>>>>>> +      struct device *dev;
> >>>>>>>> +
> >>>>>>>> +      try_module_get(THIS_MODULE);
> >>>>>>>
> >>>>>>> There should be a module_put() in pru_rproc_put()...
> >>>>>>
> >>>>>> ... and in the error path of this function.
> >>>>>>
> >>>>>>>
> >>>>>>> More comments to come tomorrow.  I'm especially worried about this API racing
> >>>>>>> with a remote processor being removed or detached.
> >>>>>>>
> >>>>>>
> >>>>>> Looking at what is done in wkup_m3_ipc_probe(), it should be possible to call
> >>>>>> rproc_get_by_handle() here and that would make sure the remote processor doesn't
> >>>>>> go away before the end of the function.
> >>>>>>
> >>>>>> More comments to come...
> >>>>
> >>>> It is possible to call rproc_get_by_handle() here instead of
> >>>> __pru_get_proc(), but that would not provide multiple functionality.
> >>>>
> >>>> The API rproc_get_by_handle() returns rproc handle on success, and NULL on
> >>>> failure where as __pru_get_proc() returns ERR_PTR on failure which provides
> >>>> multiple functionality and opportunity for us to distinguish between
> >>>> multiple errors.
> >>>>
> >>>> So we have these three options.
> >>>>
> >>>> 1. If we're using the API rproc_get_by_handle() and we want the multiple
> >>>> ERR_PTR on failure then we will need to change the API rproc_get_by_handle()
> >>>> and also all the functions that uses rproc_get_by_handle().
> >>>>
> >>>
> >>> Not optimal.
> >>>
> >>>> 2. Keep the API rproc_get_by_handle() as it is. That will restrict us from
> >>>> using multiple ERR_PTR on different kinds of error.
> >>>>
> >>>
> >>> Not optimal.
> >>>
> >>>> 3. Instead of using rproc_get_by_handle(), keep using __pru_get_proc(). This
> >>>> will make sure we have the proper ERR_PTR to retrun for different kinds of
> >>>> errors.
> >>>>
> >>>
> >>> Unacceptable for the reason I already stated.
> >>>
> >>>> Please let me know which option to continue with.
> >>>
> >>> I suggest building a wrapper that does everything you want around rproc_get_by_phandle().
>
> We can introduce a new API __rproc_get_by_phandle() similar to the API
> rproc_get_by_phandle(). The new API __rproc_get_by_phandle() will do all the
> functionality of getting the rproc. On success it will return rproc and on
> failure it will return the different ERR_PTR.
> If rproc is not probed yet, it will return ERR_PTR(-EPROBE_DEFER).
>
> This will make sure that we're getting different error codes for different
> errors from pru_rproc_get().
>
> The old API rproc_get_by_handle() will invoke the new API. On success the new
> API will return rproc and the old API will also return rproc. On failure the
> new API will return different error codes while the old API will preserve it's
> nature and return NULL.

I meant to create a wrapper around rproc_get_by_handle() that is local
to pru_rproc.c.  That way you can enact the behavior you want without
having to constrain others in this specific design.

Thanks,
Mathieu

>
> The API __rproc_get_by_handle will look like this.
>
> struct rproc *__rproc_get_by_handle(phandle phandle)
> {
>         struct rproc *rproc = NULL, *r;
>         struct device_node *np;
>
>         np = of_find_node_by_phandle(phandle);
>         if (!np)
>                 return NULL;
>
>         rcu_read_lock();
>         list_for_each_entry_rcu(r, &rproc_list, node) {
>                 if (r->dev.parent && r->dev.parent->of_node == np) {
>                         /*prevent underlying implementation from being removed */
>                         if (!try_module_get(r->dev.parent->driver->owner)) {
>                                 dev_err(&r->dev, "can't get owner\n");
>                                 break;
>                         }
>
>                         rproc = r;
>                         get_device(&rproc->dev);
>                         break;
>                 }
>         }
>         rcu_read_unlock();
>
>         of_node_put(np);
>
>         if(!rproc)
>                 return ERR_PTR(-EPROBE_DEFER);
>
>         return rproc;
> }
>
> The API rproc_get_by_handle() will look like this.
>
> struct rproc *rproc_get_by_phandle(phandle phandle)
> {
>         struct rproc *rproc = NULL;
>
>         rproc = __rproc_get_by_handle(phandle);
>
>         if(!rproc || IS_ERR(rproc))
>                 return NULL;
>
>         return rproc;
> }
>
> This way in pru_rproc_get(), we'll get the rproc by phandle and we'll still
> return different error codes depending upon failure cases. We'll also be able
> to preserve the actual functionality of rproc_get_by_phandle() so that the
> other APIs using rproc_get_by_phandle() won't get affected.
>
> Please let me know if this looks good.
>
> >>>
> >>
> >> Sure, I'll do that. I'll share this change in v6 of this patch series.
> >
> > I'm able to test the TI prueth driver from the ti-linux-5.10.y tree [1] on a
> > AM5749 cpu (custom board). But I need a more recent kernel (at least 5.15) to
> > support other devices recently added to the Linux kernel (wifi6 module and an
> > ethernet switch). Also it would be nice if this driver is finally merged in the
> > Linux kernel.
> >
> > Maybe I can help to test this series but I noticed it only provide the driver
> > for TI AM654x cpus [2]. Can you also provide patches for basic EMAC support with
> > the TI AM574x too? (I don't need advanced features like frame timestamping, HSR
> > etc).
> >
> > Also, what about patches present in the ti-linux-kernel tree and not included
> > this this series? Especially patches that modify the kernel network stack [3]
> > (net/rpmsg: add support for new rpmsg sockets). Is this new socket protocol
> > really needed?
> >
> > Notice the patch adding the rpmsg sockets [3] already conflict with the upstream
> > kernel since the AF_MCTP definition now use the value temporarly used by
> > AF_RPMSG [4].
> >
> > Can you send an updated version of the complete series?
> >
> > Thanks!
> >
> > [1] https://git.ti.com/cgit/ti-linux-kernel/ti-linux-kernel/log/?h=ti-linux-5.10.y
> >
> > [2] https://lore.kernel.org/linux-remoteproc/20220406094358.7895-1-p-mohan@ti.com/
> >
> > [3]
> > https://git.ti.com/cgit/ti-linux-kernel/ti-linux-kernel/commit/?h=ti-linux-5.10.y&id=f4b978a978c38149f712ddd137f12ed5fb914161
> >
> > [4]
> > https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=bc49d8169aa72295104f1558830c568efb946315
> >
> > Best regards,
> > Romain
> >
> >
> >>
> >>>>
> >>>> Thanks,
> >>>> Danish
> >>>>
> >>>>
> >>>>>>
> >>>>>>
> >>>>>>> Thanks,
> >>>>>>> Mathieu
> >>>>>>>
> >>>>>>>> +
> >>>>>>>> +      rproc = __pru_rproc_get(np, index);
> >>>>>>>> +      if (IS_ERR(rproc))
> >>>>>>>> +              return rproc;
> >>>>>>>> +
> >>>>>>>> +      pru = rproc->priv;
> >>>>>>>> +      dev = &rproc->dev;
> >>>>>>>> +
> >>>>>>>> +      mutex_lock(&pru->lock);
> >>>>>>>> +
> >>>>>>>> +      if (pru->client_np) {
> >>>>>>>> +              mutex_unlock(&pru->lock);
> >>>>>>>> +              put_device(dev);
> >>>>>>>> +              return ERR_PTR(-EBUSY);
> >>>>>>>> +      }
> >>>>>>>> +
> >>>>>>>> +      pru->client_np = np;
> >>>>>>>> +
> >>>>>>>> +      mutex_unlock(&pru->lock);
> >>>>>>>> +
> >>>>>>>> +      if (pru_id)
> >>>>>>>> +              *pru_id = pru->id;
> >>>>>>>> +
> >>>>>>>> +      return rproc;
> >>>>>>>> +}
> >>>>>>>> +EXPORT_SYMBOL_GPL(pru_rproc_get);
> >>>>>>>> +
> >>>>>>>> +/**
> >>>>>>>> + * pru_rproc_put() - release the PRU rproc resource
> >>>>>>>> + * @rproc: the rproc resource to release
> >>>>>>>> + *
> >>>>>>>> + * Releases the PRU rproc resource and makes it available to other
> >>>>>>>> + * users.
> >>>>>>>> + */
> >>>>>>>> +void pru_rproc_put(struct rproc *rproc)
> >>>>>>>> +{
> >>>>>>>> +      struct pru_rproc *pru;
> >>>>>>>> +
> >>>>>>>> +      if (IS_ERR_OR_NULL(rproc) || !is_pru_rproc(rproc->dev.parent))
> >>>>>>>> +              return;
> >>>>>>>> +
> >>>>>>>> +      pru = rproc->priv;
> >>>>>>>> +
> >>>>>>>> +      mutex_lock(&pru->lock);
> >>>>>>>> +
> >>>>>>>> +      if (!pru->client_np) {
> >>>>>>>> +              mutex_unlock(&pru->lock);
> >>>>>>>> +              return;
> >>>>>>>> +      }
> >>>>>>>> +
> >>>>>>>> +      pru->client_np = NULL;
> >>>>>>>> +      mutex_unlock(&pru->lock);
> >>>>>>>> +
> >>>>>>>> +      put_device(&rproc->dev);
> >>>>>>>> +}
> >>>>>>>> +EXPORT_SYMBOL_GPL(pru_rproc_put);
> >>>>>>>> +
> >>>>>>>>   static inline u32 pru_debug_read_reg(struct pru_rproc *pru, unsigned int reg)
> >>>>>>>>   {
> >>>>>>>>        return readl_relaxed(pru->mem_regions[PRU_IOMEM_DEBUG].va + reg);
> >>>>>>>> @@ -438,7 +563,7 @@ static void *pru_d_da_to_va(struct pru_rproc *pru, u32 da, size_t len)
> >>>>>>>>        dram0 = pruss->mem_regions[PRUSS_MEM_DRAM0];
> >>>>>>>>        dram1 = pruss->mem_regions[PRUSS_MEM_DRAM1];
> >>>>>>>>        /* PRU1 has its local RAM addresses reversed */
> >>>>>>>> -      if (pru->id == 1)
> >>>>>>>> +      if (pru->id == PRUSS_PRU1)
> >>>>>>>>                swap(dram0, dram1);
> >>>>>>>>        shrd_ram = pruss->mem_regions[PRUSS_MEM_SHRD_RAM2];
> >>>>>>>> @@ -747,14 +872,14 @@ static int pru_rproc_set_id(struct pru_rproc *pru)
> >>>>>>>>        case RTU0_IRAM_ADDR_MASK:
> >>>>>>>>                fallthrough;
> >>>>>>>>        case PRU0_IRAM_ADDR_MASK:
> >>>>>>>> -              pru->id = 0;
> >>>>>>>> +              pru->id = PRUSS_PRU0;
> >>>>>>>>                break;
> >>>>>>>>        case TX_PRU1_IRAM_ADDR_MASK:
> >>>>>>>>                fallthrough;
> >>>>>>>>        case RTU1_IRAM_ADDR_MASK:
> >>>>>>>>                fallthrough;
> >>>>>>>>        case PRU1_IRAM_ADDR_MASK:
> >>>>>>>> -              pru->id = 1;
> >>>>>>>> +              pru->id = PRUSS_PRU1;
> >>>>>>>>                break;
> >>>>>>>>        default:
> >>>>>>>>                ret = -EINVAL;
> >>>>>>>> @@ -816,6 +941,8 @@ static int pru_rproc_probe(struct platform_device *pdev)
> >>>>>>>>        pru->pruss = platform_get_drvdata(ppdev);
> >>>>>>>>        pru->rproc = rproc;
> >>>>>>>>        pru->fw_name = fw_name;
> >>>>>>>> +      pru->client_np = NULL;
> >>>>>>>> +      mutex_init(&pru->lock);
> >>>>>>>>        for (i = 0; i < ARRAY_SIZE(mem_names); i++) {
> >>>>>>>>                res = platform_get_resource_byname(pdev, IORESOURCE_MEM,
> >>>>>>>> @@ -903,7 +1030,7 @@ MODULE_DEVICE_TABLE(of, pru_rproc_match);
> >>>>>>>>   static struct platform_driver pru_rproc_driver = {
> >>>>>>>>        .driver = {
> >>>>>>>> -              .name   = "pru-rproc",
> >>>>>>>> +              .name   = PRU_RPROC_DRVNAME,
> >>>>>>>>                .of_match_table = pru_rproc_match,
> >>>>>>>>                .suppress_bind_attrs = true,
> >>>>>>>>        },
> >>>>>>>> @@ -915,5 +1042,6 @@ module_platform_driver(pru_rproc_driver);
> >>>>>>>>   MODULE_AUTHOR("Suman Anna <s-anna@ti.com>");
> >>>>>>>>   MODULE_AUTHOR("Andrew F. Davis <afd@ti.com>");
> >>>>>>>>   MODULE_AUTHOR("Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>");
> >>>>>>>> +MODULE_AUTHOR("Puranjay Mohan <p-mohan@ti.com>");
> >>>>>>>>   MODULE_DESCRIPTION("PRU-ICSS Remote Processor Driver");
> >>>>>>>>   MODULE_LICENSE("GPL v2");
> >>>>>>>> diff --git a/include/linux/pruss.h b/include/linux/pruss.h
> >>>>>>>> new file mode 100644
> >>>>>>>> index 000000000000..fdc719b43db0
> >>>>>>>> --- /dev/null
> >>>>>>>> +++ b/include/linux/pruss.h
> >>>>>>>> @@ -0,0 +1,56 @@
> >>>>>>>> +/* SPDX-License-Identifier: GPL-2.0-only */
> >>>>>>>> +/**
> >>>>>>>> + * PRU-ICSS Subsystem user interfaces
> >>>>>>>> + *
> >>>>>>>> + * Copyright (C) 2015-2022 Texas Instruments Incorporated - http://www.ti.com
> >>>>>>>> + *    Suman Anna <s-anna@ti.com>
> >>>>>>>> + */
> >>>>>>>> +
> >>>>>>>> +#ifndef __LINUX_PRUSS_H
> >>>>>>>> +#define __LINUX_PRUSS_H
> >>>>>>>> +
> >>>>>>>> +#include <linux/device.h>
> >>>>>>>> +#include <linux/types.h>
> >>>>>>>> +
> >>>>>>>> +#define PRU_RPROC_DRVNAME "pru-rproc"
> >>>>>>>> +
> >>>>>>>> +/*
> >>>>>>>> + * enum pruss_pru_id - PRU core identifiers
> >>>>>>>> + */
> >>>>>>>> +enum pruss_pru_id {
> >>>>>>>> +      PRUSS_PRU0 = 0,
> >>>>>>>> +      PRUSS_PRU1,
> >>>>>>>> +      PRUSS_NUM_PRUS,
> >>>>>>>> +};
> >>>>>>>> +
> >>>>>>>> +struct device_node;
> >>>>>>>> +
> >>>>>>>> +#if IS_ENABLED(CONFIG_PRU_REMOTEPROC)
> >>>>>>>> +
> >>>>>>>> +struct rproc *pru_rproc_get(struct device_node *np, int index,
> >>>>>>>> +                          enum pruss_pru_id *pru_id);
> >>>>>>>> +void pru_rproc_put(struct rproc *rproc);
> >>>>>>>> +
> >>>>>>>> +#else
> >>>>>>>> +
> >>>>>>>> +static inline struct rproc *
> >>>>>>>> +pru_rproc_get(struct device_node *np, int index, enum pruss_pru_id *pru_id)
> >>>>>>>> +{
> >>>>>>>> +      return ERR_PTR(-EOPNOTSUPP);
> >>>>>>>> +}
> >>>>>>>> +
> >>>>>>>> +static inline void pru_rproc_put(struct rproc *rproc) { }
> >>>>>>>> +
> >>>>>>>> +#endif /* CONFIG_PRU_REMOTEPROC */
> >>>>>>>> +
> >>>>>>>> +static inline bool is_pru_rproc(struct device *dev)
> >>>>>>>> +{
> >>>>>>>> +      const char *drv_name = dev_driver_string(dev);
> >>>>>>>> +
> >>>>>>>> +      if (strncmp(drv_name, PRU_RPROC_DRVNAME, sizeof(PRU_RPROC_DRVNAME)))
> >>>>>>>> +              return false;
> >>>>>>>> +
> >>>>>>>> +      return true;
> >>>>>>>> +}
> >>>>>>>> +
> >>>>>>>> +#endif /* __LINUX_PRUSS_H */
> >>>>>>>> --
> >>>>>>>> 2.17.1
> >>>>>>>>
> >>
> >

Re: [EXTERNAL] Re: [PATCH v5 2/6] remoteproc: pru: Add APIs to get and put the PRU cores

[Md Danish Anwar]

Hi Mathieu,

On 20/09/22 02:15, Mathieu Poirier wrote:
> On Tue, 13 Sept 2022 at 05:40, Md Danish Anwar <a0501179@ti.com> wrote:
>>
>> Hi Mathieu,
>>
>> On 12/09/22 18:54, Romain Naour wrote:
>>> Hi Danish, All,
>>>
>>> Le 07/09/2022 à 11:24, Md Danish Anwar a écrit :
>>>> Hi Mathieu,
>>>>
>>>> On 07/09/22 00:58, Mathieu Poirier wrote:
>>>>> On Fri, Sep 02, 2022 at 03:09:04PM +0530, Md Danish Anwar wrote:
>>>>>> Hi Mathieu,
>>>>>>
>>>>>>
>>>>>> On 15/07/22 11:52, Kishon Vijay Abraham I wrote:
>>>>>>> +Danish
>>>>>>>
>>>>>>> Hi Mathieu,
>>>>>>>
>>>>>>> On 14/07/22 22:50, Mathieu Poirier wrote:
>>>>>>>> On Tue, Jul 12, 2022 at 11:59:49AM -0600, Mathieu Poirier wrote:
>>>>>>>>> Hi Puranjay,
>>>>>>>
>>>>>>> Removed Puranjay (as he is no longer with TI) and adding Danish.
>>>>>>>
>>>>>>> Regards,
>>>>>>> Kishon
>>>>>>>>>
>>>>>>>>> On Tue, Jun 07, 2022 at 10:26:46AM +0530, Puranjay Mohan wrote:
>>>>>>>>>> From: Tero Kristo <t-kristo@ti.com>
>>>>>>>>>>
>>>>>>>>>> Add two new APIs, pru_rproc_get() and pru_rproc_put(), to the PRU
>>>>>>>>>> driver to allow client drivers to acquire and release the remoteproc
>>>>>>>>>> device associated with a PRU core. The PRU cores are treated as
>>>>>>>>>> resources with only one client owning it at a time.
>>>>>>>>>>
>>>>>>>>>> The pru_rproc_get() function returns the rproc handle corresponding
>>>>>>>>>> to a PRU core identified by the device tree "ti,prus" property under
>>>>>>>>>> the client node. The pru_rproc_put() is the complementary function
>>>>>>>>>> to pru_rproc_get().
>>>>>>>>>>
>>>>>>>>>> Co-developed-by: Suman Anna <s-anna@ti.com>
>>>>>>>>>> Signed-off-by: Suman Anna <s-anna@ti.com>
>>>>>>>>>> Signed-off-by: Tero Kristo <t-kristo@ti.com>
>>>>>>>>>> Co-developed-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
>>>>>>>>>> Signed-off-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
>>>>>>>>>> Co-developed-by: Puranjay Mohan <p-mohan@ti.com>
>>>>>>>>>> Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
>>>>>>>>>> ---
>>>>>>>>>>   drivers/remoteproc/pru_rproc.c | 138 +++++++++++++++++++++++++++++++--
>>>>>>>>>>   include/linux/pruss.h          |  56 +++++++++++++
>>>>>>>>>>   2 files changed, 189 insertions(+), 5 deletions(-)
>>>>>>>>>>   create mode 100644 include/linux/pruss.h
>>>>>>>>>>
>>>>>>>>>> diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
>>>>>>>>>> index 1777a01fa84e..7a35b400287a 100644
>>>>>>>>>> --- a/drivers/remoteproc/pru_rproc.c
>>>>>>>>>> +++ b/drivers/remoteproc/pru_rproc.c
>>>>>>>>>> @@ -2,12 +2,13 @@
>>>>>>>>>>   /*
>>>>>>>>>>    * PRU-ICSS remoteproc driver for various TI SoCs
>>>>>>>>>>    *
>>>>>>>>>> - * Copyright (C) 2014-2020 Texas Instruments Incorporated - https://www.ti.com/
>>>>>>>>>> + * Copyright (C) 2014-2022 Texas Instruments Incorporated - https://www.ti.com/
>>>>>>>>>>    *
>>>>>>>>>>    * Author(s):
>>>>>>>>>>    *   Suman Anna <s-anna@ti.com>
>>>>>>>>>>    *   Andrew F. Davis <afd@ti.com>
>>>>>>>>>>    *   Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org> for Texas Instruments
>>>>>>>>>> + *    Puranjay Mohan <p-mohan@ti.com>
>>>>>>>>>>    */
>>>>>>>>>>   #include <linux/bitops.h>
>>>>>>>>>> @@ -16,6 +17,7 @@
>>>>>>>>>>   #include <linux/module.h>
>>>>>>>>>>   #include <linux/of_device.h>
>>>>>>>>>>   #include <linux/of_irq.h>
>>>>>>>>>> +#include <linux/pruss.h>
>>>>>>>>>>   #include <linux/pruss_driver.h>
>>>>>>>>>>   #include <linux/remoteproc.h>
>>>>>>>>>> @@ -111,6 +113,8 @@ struct pru_private_data {
>>>>>>>>>>    * @rproc: remoteproc pointer for this PRU core
>>>>>>>>>>    * @data: PRU core specific data
>>>>>>>>>>    * @mem_regions: data for each of the PRU memory regions
>>>>>>>>>> + * @client_np: client device node
>>>>>>>>>> + * @lock: mutex to protect client usage
>>>>>>>>>>    * @fw_name: name of firmware image used during loading
>>>>>>>>>>    * @mapped_irq: virtual interrupt numbers of created fw specific mapping
>>>>>>>>>>    * @pru_interrupt_map: pointer to interrupt mapping description (firmware)
>>>>>>>>>> @@ -126,6 +130,8 @@ struct pru_rproc {
>>>>>>>>>>        struct rproc *rproc;
>>>>>>>>>>        const struct pru_private_data *data;
>>>>>>>>>>        struct pruss_mem_region mem_regions[PRU_IOMEM_MAX];
>>>>>>>>>> +      struct device_node *client_np;
>>>>>>>>>> +      struct mutex lock; /* client access lock */
>>>>>>>>>>        const char *fw_name;
>>>>>>>>>>        unsigned int *mapped_irq;
>>>>>>>>>>        struct pru_irq_rsc *pru_interrupt_map;
>>>>>>>>>> @@ -146,6 +152,125 @@ void pru_control_write_reg(struct pru_rproc *pru, unsigned int reg, u32 val)
>>>>>>>>>>        writel_relaxed(val, pru->mem_regions[PRU_IOMEM_CTRL].va + reg);
>>>>>>>>>>   }
>>>>>>>>>> +static struct rproc *__pru_rproc_get(struct device_node *np, int index)
>>>>>>>>>> +{
>>>>>>>>>> +      struct device_node *rproc_np = NULL;
>>>>>>>>>> +      struct platform_device *pdev;
>>>>>>>>>> +      struct rproc *rproc;
>>>>>>>>>> +
>>>>>>>>>> +      rproc_np = of_parse_phandle(np, "ti,prus", index);
>>>>>>>>>> +      if (!rproc_np || !of_device_is_available(rproc_np))
>>>>>>>>>> +              return ERR_PTR(-ENODEV);
>>>>>>>>>> +
>>>>>>>>>> +      pdev = of_find_device_by_node(rproc_np);
>>>>>>>>>> +      of_node_put(rproc_np);
>>>>>>>>>> +
>>>>>>>>>> +      if (!pdev || !(&pdev->dev) || !((&pdev->dev)->driver))
>>>>>>>>>> +              /* probably PRU not yet probed */
>>>>>>>>>> +              return ERR_PTR(-EPROBE_DEFER);
>>>>>>>>>> +
>>>>>>>>>> +      /* make sure it is PRU rproc */
>>>>>>>>>> +      if (!is_pru_rproc(&pdev->dev)) {
>>>>>>>>>> +              put_device(&pdev->dev);
>>>>>>>>>> +              return ERR_PTR(-ENODEV);
>>>>>>>>>> +      }
>>>>>>>>>> +
>>>>>>>>>> +      rproc = platform_get_drvdata(pdev);
>>>>>>>>>> +      put_device(&pdev->dev);
>>>>>>>>>> +      if (!rproc)
>>>>>>>>>> +              return ERR_PTR(-EPROBE_DEFER);
>>>>>>>>>> +
>>>>>>>>>> +      get_device(&rproc->dev);
>>>>>>>>>> +
>>>>>>>>>> +      return rproc;
>>>>>>>>>> +}
>>>>>>>>>> +
>>>>>>>>>> +/**
>>>>>>>>>> + * pru_rproc_get() - get the PRU rproc instance from a device node
>>>>>>>>>> + * @np: the user/client device node
>>>>>>>>>> + * @index: index to use for the ti,prus property
>>>>>>>>>> + * @pru_id: optional pointer to return the PRU remoteproc processor id
>>>>>>>>>> + *
>>>>>>>>>> + * This function looks through a client device node's "ti,prus" property at
>>>>>>>>>> + * index @index and returns the rproc handle for a valid PRU remote processor if
>>>>>>>>>> + * found. The function allows only one user to own the PRU rproc resource at a
>>>>>>>>>> + * time. Caller must call pru_rproc_put() when done with using the rproc, not
>>>>>>>>>> + * required if the function returns a failure.
>>>>>>>>>> + *
>>>>>>>>>> + * When optional @pru_id pointer is passed the PRU remoteproc processor id is
>>>>>>>>>> + * returned.
>>>>>>>>>> + *
>>>>>>>>>> + * Return: rproc handle on success, and an ERR_PTR on failure using one
>>>>>>>>>> + * of the following error values
>>>>>>>>>> + *    -ENODEV if device is not found
>>>>>>>>>> + *    -EBUSY if PRU is already acquired by anyone
>>>>>>>>>> + *    -EPROBE_DEFER is PRU device is not probed yet
>>>>>>>>>> + */
>>>>>>>>>> +struct rproc *pru_rproc_get(struct device_node *np, int index,
>>>>>>>>>> +                          enum pruss_pru_id *pru_id)
>>>>>>>>>> +{
>>>>>>>>>> +      struct rproc *rproc;
>>>>>>>>>> +      struct pru_rproc *pru;
>>>>>>>>>> +      struct device *dev;
>>>>>>>>>> +
>>>>>>>>>> +      try_module_get(THIS_MODULE);
>>>>>>>>>
>>>>>>>>> There should be a module_put() in pru_rproc_put()...
>>>>>>>>
>>>>>>>> ... and in the error path of this function.
>>>>>>>>
>>>>>>>>>
>>>>>>>>> More comments to come tomorrow.  I'm especially worried about this API racing
>>>>>>>>> with a remote processor being removed or detached.
>>>>>>>>>
>>>>>>>>
>>>>>>>> Looking at what is done in wkup_m3_ipc_probe(), it should be possible to call
>>>>>>>> rproc_get_by_handle() here and that would make sure the remote processor doesn't
>>>>>>>> go away before the end of the function.
>>>>>>>>
>>>>>>>> More comments to come...
>>>>>>
>>>>>> It is possible to call rproc_get_by_handle() here instead of
>>>>>> __pru_get_proc(), but that would not provide multiple functionality.
>>>>>>
>>>>>> The API rproc_get_by_handle() returns rproc handle on success, and NULL on
>>>>>> failure where as __pru_get_proc() returns ERR_PTR on failure which provides
>>>>>> multiple functionality and opportunity for us to distinguish between
>>>>>> multiple errors.
>>>>>>
>>>>>> So we have these three options.
>>>>>>
>>>>>> 1. If we're using the API rproc_get_by_handle() and we want the multiple
>>>>>> ERR_PTR on failure then we will need to change the API rproc_get_by_handle()
>>>>>> and also all the functions that uses rproc_get_by_handle().
>>>>>>
>>>>>
>>>>> Not optimal.
>>>>>
>>>>>> 2. Keep the API rproc_get_by_handle() as it is. That will restrict us from
>>>>>> using multiple ERR_PTR on different kinds of error.
>>>>>>
>>>>>
>>>>> Not optimal.
>>>>>
>>>>>> 3. Instead of using rproc_get_by_handle(), keep using __pru_get_proc(). This
>>>>>> will make sure we have the proper ERR_PTR to retrun for different kinds of
>>>>>> errors.
>>>>>>
>>>>>
>>>>> Unacceptable for the reason I already stated.
>>>>>
>>>>>> Please let me know which option to continue with.
>>>>>
>>>>> I suggest building a wrapper that does everything you want around rproc_get_by_phandle().
>>
>> We can introduce a new API __rproc_get_by_phandle() similar to the API
>> rproc_get_by_phandle(). The new API __rproc_get_by_phandle() will do all the
>> functionality of getting the rproc. On success it will return rproc and on
>> failure it will return the different ERR_PTR.
>> If rproc is not probed yet, it will return ERR_PTR(-EPROBE_DEFER).
>>
>> This will make sure that we're getting different error codes for different
>> errors from pru_rproc_get().
>>
>> The old API rproc_get_by_handle() will invoke the new API. On success the new
>> API will return rproc and the old API will also return rproc. On failure the
>> new API will return different error codes while the old API will preserve it's
>> nature and return NULL.
> 
> I meant to create a wrapper around rproc_get_by_handle() that is local
> to pru_rproc.c.  That way you can enact the behavior you want without
> having to constrain others in this specific design.
> 
> Thanks,
> Mathieu
> 
As The API rproc_get_by_phandle() returns null on failure, I am not able to
create a wrapper around this API, which is local to pru_rproc.c to enact the
different behaviors that I want. As the API is only returning null on failure
but pru_rproc_get needs different error codes for different failures. We can
create a wrapper but that won't be able to provide different error codes that
we want in pru_rproc_get().

This is why I am suggesting to introduce new API [__rproc_get_by_phandle()]
similar to rproc_get_by_handle(). The new API can return different error codes
on different error. While the older API [rproc_get_by_handle()] will call the
new API and on failure on new API the older API will just return null
irrespective of the error code returned by the new API. Thus preserving it's
original design.

This way we can enact the behavior that we want by calling the new API while
the other APIs dependent on the older API, can still call
rproc_get_by_phandle() without any change to the existing design.

We can have a little discussion to discuss this further on some IRC channel or
meeting if it's possible.

Thanks,
Danish


>>
>> The API __rproc_get_by_handle will look like this.
>>
>> struct rproc *__rproc_get_by_handle(phandle phandle)
>> {
>>         struct rproc *rproc = NULL, *r;
>>         struct device_node *np;
>>
>>         np = of_find_node_by_phandle(phandle);
>>         if (!np)
>>                 return NULL;
>>
>>         rcu_read_lock();
>>         list_for_each_entry_rcu(r, &rproc_list, node) {
>>                 if (r->dev.parent && r->dev.parent->of_node == np) {
>>                         /*prevent underlying implementation from being removed */
>>                         if (!try_module_get(r->dev.parent->driver->owner)) {
>>                                 dev_err(&r->dev, "can't get owner\n");
>>                                 break;
>>                         }
>>
>>                         rproc = r;
>>                         get_device(&rproc->dev);
>>                         break;
>>                 }
>>         }
>>         rcu_read_unlock();
>>
>>         of_node_put(np);
>>
>>         if(!rproc)
>>                 return ERR_PTR(-EPROBE_DEFER);
>>
>>         return rproc;
>> }
>>
>> The API rproc_get_by_handle() will look like this.
>>
>> struct rproc *rproc_get_by_phandle(phandle phandle)
>> {
>>         struct rproc *rproc = NULL;
>>
>>         rproc = __rproc_get_by_handle(phandle);
>>
>>         if(!rproc || IS_ERR(rproc))
>>                 return NULL;
>>
>>         return rproc;
>> }
>>
>> This way in pru_rproc_get(), we'll get the rproc by phandle and we'll still
>> return different error codes depending upon failure cases. We'll also be able
>> to preserve the actual functionality of rproc_get_by_phandle() so that the
>> other APIs using rproc_get_by_phandle() won't get affected.
>>
>> Please let me know if this looks good.
>>
>>>>>
>>>>
>>>> Sure, I'll do that. I'll share this change in v6 of this patch series.
>>>
>>> I'm able to test the TI prueth driver from the ti-linux-5.10.y tree [1] on a
>>> AM5749 cpu (custom board). But I need a more recent kernel (at least 5.15) to
>>> support other devices recently added to the Linux kernel (wifi6 module and an
>>> ethernet switch). Also it would be nice if this driver is finally merged in the
>>> Linux kernel.
>>>
>>> Maybe I can help to test this series but I noticed it only provide the driver
>>> for TI AM654x cpus [2]. Can you also provide patches for basic EMAC support with
>>> the TI AM574x too? (I don't need advanced features like frame timestamping, HSR
>>> etc).
>>>
>>> Also, what about patches present in the ti-linux-kernel tree and not included
>>> this this series? Especially patches that modify the kernel network stack [3]
>>> (net/rpmsg: add support for new rpmsg sockets). Is this new socket protocol
>>> really needed?
>>>
>>> Notice the patch adding the rpmsg sockets [3] already conflict with the upstream
>>> kernel since the AF_MCTP definition now use the value temporarly used by
>>> AF_RPMSG [4].
>>>
>>> Can you send an updated version of the complete series?
>>>
>>> Thanks!
>>>
>>> [1] https://git.ti.com/cgit/ti-linux-kernel/ti-linux-kernel/log/?h=ti-linux-5.10.y
>>>
>>> [2] https://lore.kernel.org/linux-remoteproc/20220406094358.7895-1-p-mohan@ti.com/
>>>
>>> [3]
>>> https://git.ti.com/cgit/ti-linux-kernel/ti-linux-kernel/commit/?h=ti-linux-5.10.y&id=f4b978a978c38149f712ddd137f12ed5fb914161
>>>
>>> [4]
>>> https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=bc49d8169aa72295104f1558830c568efb946315
>>>
>>> Best regards,
>>> Romain
>>>
>>>
>>>>
>>>>>>
>>>>>> Thanks,
>>>>>> Danish
>>>>>>
>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>> Thanks,
>>>>>>>>> Mathieu
>>>>>>>>>
>>>>>>>>>> +
>>>>>>>>>> +      rproc = __pru_rproc_get(np, index);
>>>>>>>>>> +      if (IS_ERR(rproc))
>>>>>>>>>> +              return rproc;
>>>>>>>>>> +
>>>>>>>>>> +      pru = rproc->priv;
>>>>>>>>>> +      dev = &rproc->dev;
>>>>>>>>>> +
>>>>>>>>>> +      mutex_lock(&pru->lock);
>>>>>>>>>> +
>>>>>>>>>> +      if (pru->client_np) {
>>>>>>>>>> +              mutex_unlock(&pru->lock);
>>>>>>>>>> +              put_device(dev);
>>>>>>>>>> +              return ERR_PTR(-EBUSY);
>>>>>>>>>> +      }
>>>>>>>>>> +
>>>>>>>>>> +      pru->client_np = np;
>>>>>>>>>> +
>>>>>>>>>> +      mutex_unlock(&pru->lock);
>>>>>>>>>> +
>>>>>>>>>> +      if (pru_id)
>>>>>>>>>> +              *pru_id = pru->id;
>>>>>>>>>> +
>>>>>>>>>> +      return rproc;
>>>>>>>>>> +}
>>>>>>>>>> +EXPORT_SYMBOL_GPL(pru_rproc_get);
>>>>>>>>>> +
>>>>>>>>>> +/**
>>>>>>>>>> + * pru_rproc_put() - release the PRU rproc resource
>>>>>>>>>> + * @rproc: the rproc resource to release
>>>>>>>>>> + *
>>>>>>>>>> + * Releases the PRU rproc resource and makes it available to other
>>>>>>>>>> + * users.
>>>>>>>>>> + */
>>>>>>>>>> +void pru_rproc_put(struct rproc *rproc)
>>>>>>>>>> +{
>>>>>>>>>> +      struct pru_rproc *pru;
>>>>>>>>>> +
>>>>>>>>>> +      if (IS_ERR_OR_NULL(rproc) || !is_pru_rproc(rproc->dev.parent))
>>>>>>>>>> +              return;
>>>>>>>>>> +
>>>>>>>>>> +      pru = rproc->priv;
>>>>>>>>>> +
>>>>>>>>>> +      mutex_lock(&pru->lock);
>>>>>>>>>> +
>>>>>>>>>> +      if (!pru->client_np) {
>>>>>>>>>> +              mutex_unlock(&pru->lock);
>>>>>>>>>> +              return;
>>>>>>>>>> +      }
>>>>>>>>>> +
>>>>>>>>>> +      pru->client_np = NULL;
>>>>>>>>>> +      mutex_unlock(&pru->lock);
>>>>>>>>>> +
>>>>>>>>>> +      put_device(&rproc->dev);
>>>>>>>>>> +}
>>>>>>>>>> +EXPORT_SYMBOL_GPL(pru_rproc_put);
>>>>>>>>>> +
>>>>>>>>>>   static inline u32 pru_debug_read_reg(struct pru_rproc *pru, unsigned int reg)
>>>>>>>>>>   {
>>>>>>>>>>        return readl_relaxed(pru->mem_regions[PRU_IOMEM_DEBUG].va + reg);
>>>>>>>>>> @@ -438,7 +563,7 @@ static void *pru_d_da_to_va(struct pru_rproc *pru, u32 da, size_t len)
>>>>>>>>>>        dram0 = pruss->mem_regions[PRUSS_MEM_DRAM0];
>>>>>>>>>>        dram1 = pruss->mem_regions[PRUSS_MEM_DRAM1];
>>>>>>>>>>        /* PRU1 has its local RAM addresses reversed */
>>>>>>>>>> -      if (pru->id == 1)
>>>>>>>>>> +      if (pru->id == PRUSS_PRU1)
>>>>>>>>>>                swap(dram0, dram1);
>>>>>>>>>>        shrd_ram = pruss->mem_regions[PRUSS_MEM_SHRD_RAM2];
>>>>>>>>>> @@ -747,14 +872,14 @@ static int pru_rproc_set_id(struct pru_rproc *pru)
>>>>>>>>>>        case RTU0_IRAM_ADDR_MASK:
>>>>>>>>>>                fallthrough;
>>>>>>>>>>        case PRU0_IRAM_ADDR_MASK:
>>>>>>>>>> -              pru->id = 0;
>>>>>>>>>> +              pru->id = PRUSS_PRU0;
>>>>>>>>>>                break;
>>>>>>>>>>        case TX_PRU1_IRAM_ADDR_MASK:
>>>>>>>>>>                fallthrough;
>>>>>>>>>>        case RTU1_IRAM_ADDR_MASK:
>>>>>>>>>>                fallthrough;
>>>>>>>>>>        case PRU1_IRAM_ADDR_MASK:
>>>>>>>>>> -              pru->id = 1;
>>>>>>>>>> +              pru->id = PRUSS_PRU1;
>>>>>>>>>>                break;
>>>>>>>>>>        default:
>>>>>>>>>>                ret = -EINVAL;
>>>>>>>>>> @@ -816,6 +941,8 @@ static int pru_rproc_probe(struct platform_device *pdev)
>>>>>>>>>>        pru->pruss = platform_get_drvdata(ppdev);
>>>>>>>>>>        pru->rproc = rproc;
>>>>>>>>>>        pru->fw_name = fw_name;
>>>>>>>>>> +      pru->client_np = NULL;
>>>>>>>>>> +      mutex_init(&pru->lock);
>>>>>>>>>>        for (i = 0; i < ARRAY_SIZE(mem_names); i++) {
>>>>>>>>>>                res = platform_get_resource_byname(pdev, IORESOURCE_MEM,
>>>>>>>>>> @@ -903,7 +1030,7 @@ MODULE_DEVICE_TABLE(of, pru_rproc_match);
>>>>>>>>>>   static struct platform_driver pru_rproc_driver = {
>>>>>>>>>>        .driver = {
>>>>>>>>>> -              .name   = "pru-rproc",
>>>>>>>>>> +              .name   = PRU_RPROC_DRVNAME,
>>>>>>>>>>                .of_match_table = pru_rproc_match,
>>>>>>>>>>                .suppress_bind_attrs = true,
>>>>>>>>>>        },
>>>>>>>>>> @@ -915,5 +1042,6 @@ module_platform_driver(pru_rproc_driver);
>>>>>>>>>>   MODULE_AUTHOR("Suman Anna <s-anna@ti.com>");
>>>>>>>>>>   MODULE_AUTHOR("Andrew F. Davis <afd@ti.com>");
>>>>>>>>>>   MODULE_AUTHOR("Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>");
>>>>>>>>>> +MODULE_AUTHOR("Puranjay Mohan <p-mohan@ti.com>");
>>>>>>>>>>   MODULE_DESCRIPTION("PRU-ICSS Remote Processor Driver");
>>>>>>>>>>   MODULE_LICENSE("GPL v2");
>>>>>>>>>> diff --git a/include/linux/pruss.h b/include/linux/pruss.h
>>>>>>>>>> new file mode 100644
>>>>>>>>>> index 000000000000..fdc719b43db0
>>>>>>>>>> --- /dev/null
>>>>>>>>>> +++ b/include/linux/pruss.h
>>>>>>>>>> @@ -0,0 +1,56 @@
>>>>>>>>>> +/* SPDX-License-Identifier: GPL-2.0-only */
>>>>>>>>>> +/**
>>>>>>>>>> + * PRU-ICSS Subsystem user interfaces
>>>>>>>>>> + *
>>>>>>>>>> + * Copyright (C) 2015-2022 Texas Instruments Incorporated - http://www.ti.com
>>>>>>>>>> + *    Suman Anna <s-anna@ti.com>
>>>>>>>>>> + */
>>>>>>>>>> +
>>>>>>>>>> +#ifndef __LINUX_PRUSS_H
>>>>>>>>>> +#define __LINUX_PRUSS_H
>>>>>>>>>> +
>>>>>>>>>> +#include <linux/device.h>
>>>>>>>>>> +#include <linux/types.h>
>>>>>>>>>> +
>>>>>>>>>> +#define PRU_RPROC_DRVNAME "pru-rproc"
>>>>>>>>>> +
>>>>>>>>>> +/*
>>>>>>>>>> + * enum pruss_pru_id - PRU core identifiers
>>>>>>>>>> + */
>>>>>>>>>> +enum pruss_pru_id {
>>>>>>>>>> +      PRUSS_PRU0 = 0,
>>>>>>>>>> +      PRUSS_PRU1,
>>>>>>>>>> +      PRUSS_NUM_PRUS,
>>>>>>>>>> +};
>>>>>>>>>> +
>>>>>>>>>> +struct device_node;
>>>>>>>>>> +
>>>>>>>>>> +#if IS_ENABLED(CONFIG_PRU_REMOTEPROC)
>>>>>>>>>> +
>>>>>>>>>> +struct rproc *pru_rproc_get(struct device_node *np, int index,
>>>>>>>>>> +                          enum pruss_pru_id *pru_id);
>>>>>>>>>> +void pru_rproc_put(struct rproc *rproc);
>>>>>>>>>> +
>>>>>>>>>> +#else
>>>>>>>>>> +
>>>>>>>>>> +static inline struct rproc *
>>>>>>>>>> +pru_rproc_get(struct device_node *np, int index, enum pruss_pru_id *pru_id)
>>>>>>>>>> +{
>>>>>>>>>> +      return ERR_PTR(-EOPNOTSUPP);
>>>>>>>>>> +}
>>>>>>>>>> +
>>>>>>>>>> +static inline void pru_rproc_put(struct rproc *rproc) { }
>>>>>>>>>> +
>>>>>>>>>> +#endif /* CONFIG_PRU_REMOTEPROC */
>>>>>>>>>> +
>>>>>>>>>> +static inline bool is_pru_rproc(struct device *dev)
>>>>>>>>>> +{
>>>>>>>>>> +      const char *drv_name = dev_driver_string(dev);
>>>>>>>>>> +
>>>>>>>>>> +      if (strncmp(drv_name, PRU_RPROC_DRVNAME, sizeof(PRU_RPROC_DRVNAME)))
>>>>>>>>>> +              return false;
>>>>>>>>>> +
>>>>>>>>>> +      return true;
>>>>>>>>>> +}
>>>>>>>>>> +
>>>>>>>>>> +#endif /* __LINUX_PRUSS_H */
>>>>>>>>>> --
>>>>>>>>>> 2.17.1
>>>>>>>>>>
>>>>
>>>

Re: [EXTERNAL] Re: [PATCH v5 2/6] remoteproc: pru: Add APIs to get and put the PRU cores

[Mathieu Poirier]

On Mon, Sep 26, 2022 at 03:53:09PM +0530, Md Danish Anwar wrote:
> Hi Mathieu,
> 
> On 20/09/22 02:15, Mathieu Poirier wrote:
> > On Tue, 13 Sept 2022 at 05:40, Md Danish Anwar <a0501179@ti.com> wrote:
> >>
> >> Hi Mathieu,
> >>
> >> On 12/09/22 18:54, Romain Naour wrote:
> >>> Hi Danish, All,
> >>>
> >>> Le 07/09/2022 à 11:24, Md Danish Anwar a écrit :
> >>>> Hi Mathieu,
> >>>>
> >>>> On 07/09/22 00:58, Mathieu Poirier wrote:
> >>>>> On Fri, Sep 02, 2022 at 03:09:04PM +0530, Md Danish Anwar wrote:
> >>>>>> Hi Mathieu,
> >>>>>>
> >>>>>>
> >>>>>> On 15/07/22 11:52, Kishon Vijay Abraham I wrote:
> >>>>>>> +Danish
> >>>>>>>
> >>>>>>> Hi Mathieu,
> >>>>>>>
> >>>>>>> On 14/07/22 22:50, Mathieu Poirier wrote:
> >>>>>>>> On Tue, Jul 12, 2022 at 11:59:49AM -0600, Mathieu Poirier wrote:
> >>>>>>>>> Hi Puranjay,
> >>>>>>>
> >>>>>>> Removed Puranjay (as he is no longer with TI) and adding Danish.
> >>>>>>>
> >>>>>>> Regards,
> >>>>>>> Kishon
> >>>>>>>>>
> >>>>>>>>> On Tue, Jun 07, 2022 at 10:26:46AM +0530, Puranjay Mohan wrote:
> >>>>>>>>>> From: Tero Kristo <t-kristo@ti.com>
> >>>>>>>>>>
> >>>>>>>>>> Add two new APIs, pru_rproc_get() and pru_rproc_put(), to the PRU
> >>>>>>>>>> driver to allow client drivers to acquire and release the remoteproc
> >>>>>>>>>> device associated with a PRU core. The PRU cores are treated as
> >>>>>>>>>> resources with only one client owning it at a time.
> >>>>>>>>>>
> >>>>>>>>>> The pru_rproc_get() function returns the rproc handle corresponding
> >>>>>>>>>> to a PRU core identified by the device tree "ti,prus" property under
> >>>>>>>>>> the client node. The pru_rproc_put() is the complementary function
> >>>>>>>>>> to pru_rproc_get().
> >>>>>>>>>>
> >>>>>>>>>> Co-developed-by: Suman Anna <s-anna@ti.com>
> >>>>>>>>>> Signed-off-by: Suman Anna <s-anna@ti.com>
> >>>>>>>>>> Signed-off-by: Tero Kristo <t-kristo@ti.com>
> >>>>>>>>>> Co-developed-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
> >>>>>>>>>> Signed-off-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
> >>>>>>>>>> Co-developed-by: Puranjay Mohan <p-mohan@ti.com>
> >>>>>>>>>> Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
> >>>>>>>>>> ---
> >>>>>>>>>>   drivers/remoteproc/pru_rproc.c | 138 +++++++++++++++++++++++++++++++--
> >>>>>>>>>>   include/linux/pruss.h          |  56 +++++++++++++
> >>>>>>>>>>   2 files changed, 189 insertions(+), 5 deletions(-)
> >>>>>>>>>>   create mode 100644 include/linux/pruss.h
> >>>>>>>>>>
> >>>>>>>>>> diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
> >>>>>>>>>> index 1777a01fa84e..7a35b400287a 100644
> >>>>>>>>>> --- a/drivers/remoteproc/pru_rproc.c
> >>>>>>>>>> +++ b/drivers/remoteproc/pru_rproc.c
> >>>>>>>>>> @@ -2,12 +2,13 @@
> >>>>>>>>>>   /*
> >>>>>>>>>>    * PRU-ICSS remoteproc driver for various TI SoCs
> >>>>>>>>>>    *
> >>>>>>>>>> - * Copyright (C) 2014-2020 Texas Instruments Incorporated - https://www.ti.com/
> >>>>>>>>>> + * Copyright (C) 2014-2022 Texas Instruments Incorporated - https://www.ti.com/
> >>>>>>>>>>    *
> >>>>>>>>>>    * Author(s):
> >>>>>>>>>>    *   Suman Anna <s-anna@ti.com>
> >>>>>>>>>>    *   Andrew F. Davis <afd@ti.com>
> >>>>>>>>>>    *   Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org> for Texas Instruments
> >>>>>>>>>> + *    Puranjay Mohan <p-mohan@ti.com>
> >>>>>>>>>>    */
> >>>>>>>>>>   #include <linux/bitops.h>
> >>>>>>>>>> @@ -16,6 +17,7 @@
> >>>>>>>>>>   #include <linux/module.h>
> >>>>>>>>>>   #include <linux/of_device.h>
> >>>>>>>>>>   #include <linux/of_irq.h>
> >>>>>>>>>> +#include <linux/pruss.h>
> >>>>>>>>>>   #include <linux/pruss_driver.h>
> >>>>>>>>>>   #include <linux/remoteproc.h>
> >>>>>>>>>> @@ -111,6 +113,8 @@ struct pru_private_data {
> >>>>>>>>>>    * @rproc: remoteproc pointer for this PRU core
> >>>>>>>>>>    * @data: PRU core specific data
> >>>>>>>>>>    * @mem_regions: data for each of the PRU memory regions
> >>>>>>>>>> + * @client_np: client device node
> >>>>>>>>>> + * @lock: mutex to protect client usage
> >>>>>>>>>>    * @fw_name: name of firmware image used during loading
> >>>>>>>>>>    * @mapped_irq: virtual interrupt numbers of created fw specific mapping
> >>>>>>>>>>    * @pru_interrupt_map: pointer to interrupt mapping description (firmware)
> >>>>>>>>>> @@ -126,6 +130,8 @@ struct pru_rproc {
> >>>>>>>>>>        struct rproc *rproc;
> >>>>>>>>>>        const struct pru_private_data *data;
> >>>>>>>>>>        struct pruss_mem_region mem_regions[PRU_IOMEM_MAX];
> >>>>>>>>>> +      struct device_node *client_np;
> >>>>>>>>>> +      struct mutex lock; /* client access lock */
> >>>>>>>>>>        const char *fw_name;
> >>>>>>>>>>        unsigned int *mapped_irq;
> >>>>>>>>>>        struct pru_irq_rsc *pru_interrupt_map;
> >>>>>>>>>> @@ -146,6 +152,125 @@ void pru_control_write_reg(struct pru_rproc *pru, unsigned int reg, u32 val)
> >>>>>>>>>>        writel_relaxed(val, pru->mem_regions[PRU_IOMEM_CTRL].va + reg);
> >>>>>>>>>>   }
> >>>>>>>>>> +static struct rproc *__pru_rproc_get(struct device_node *np, int index)
> >>>>>>>>>> +{
> >>>>>>>>>> +      struct device_node *rproc_np = NULL;
> >>>>>>>>>> +      struct platform_device *pdev;
> >>>>>>>>>> +      struct rproc *rproc;
> >>>>>>>>>> +
> >>>>>>>>>> +      rproc_np = of_parse_phandle(np, "ti,prus", index);
> >>>>>>>>>> +      if (!rproc_np || !of_device_is_available(rproc_np))
> >>>>>>>>>> +              return ERR_PTR(-ENODEV);
> >>>>>>>>>> +
> >>>>>>>>>> +      pdev = of_find_device_by_node(rproc_np);
> >>>>>>>>>> +      of_node_put(rproc_np);
> >>>>>>>>>> +
> >>>>>>>>>> +      if (!pdev || !(&pdev->dev) || !((&pdev->dev)->driver))
> >>>>>>>>>> +              /* probably PRU not yet probed */
> >>>>>>>>>> +              return ERR_PTR(-EPROBE_DEFER);
> >>>>>>>>>> +
> >>>>>>>>>> +      /* make sure it is PRU rproc */
> >>>>>>>>>> +      if (!is_pru_rproc(&pdev->dev)) {
> >>>>>>>>>> +              put_device(&pdev->dev);
> >>>>>>>>>> +              return ERR_PTR(-ENODEV);
> >>>>>>>>>> +      }
> >>>>>>>>>> +
> >>>>>>>>>> +      rproc = platform_get_drvdata(pdev);
> >>>>>>>>>> +      put_device(&pdev->dev);
> >>>>>>>>>> +      if (!rproc)
> >>>>>>>>>> +              return ERR_PTR(-EPROBE_DEFER);
> >>>>>>>>>> +
> >>>>>>>>>> +      get_device(&rproc->dev);
> >>>>>>>>>> +
> >>>>>>>>>> +      return rproc;
> >>>>>>>>>> +}
> >>>>>>>>>> +
> >>>>>>>>>> +/**
> >>>>>>>>>> + * pru_rproc_get() - get the PRU rproc instance from a device node
> >>>>>>>>>> + * @np: the user/client device node
> >>>>>>>>>> + * @index: index to use for the ti,prus property
> >>>>>>>>>> + * @pru_id: optional pointer to return the PRU remoteproc processor id
> >>>>>>>>>> + *
> >>>>>>>>>> + * This function looks through a client device node's "ti,prus" property at
> >>>>>>>>>> + * index @index and returns the rproc handle for a valid PRU remote processor if
> >>>>>>>>>> + * found. The function allows only one user to own the PRU rproc resource at a
> >>>>>>>>>> + * time. Caller must call pru_rproc_put() when done with using the rproc, not
> >>>>>>>>>> + * required if the function returns a failure.
> >>>>>>>>>> + *
> >>>>>>>>>> + * When optional @pru_id pointer is passed the PRU remoteproc processor id is
> >>>>>>>>>> + * returned.
> >>>>>>>>>> + *
> >>>>>>>>>> + * Return: rproc handle on success, and an ERR_PTR on failure using one
> >>>>>>>>>> + * of the following error values
> >>>>>>>>>> + *    -ENODEV if device is not found
> >>>>>>>>>> + *    -EBUSY if PRU is already acquired by anyone
> >>>>>>>>>> + *    -EPROBE_DEFER is PRU device is not probed yet
> >>>>>>>>>> + */
> >>>>>>>>>> +struct rproc *pru_rproc_get(struct device_node *np, int index,
> >>>>>>>>>> +                          enum pruss_pru_id *pru_id)
> >>>>>>>>>> +{
> >>>>>>>>>> +      struct rproc *rproc;
> >>>>>>>>>> +      struct pru_rproc *pru;
> >>>>>>>>>> +      struct device *dev;
> >>>>>>>>>> +
> >>>>>>>>>> +      try_module_get(THIS_MODULE);
> >>>>>>>>>
> >>>>>>>>> There should be a module_put() in pru_rproc_put()...
> >>>>>>>>
> >>>>>>>> ... and in the error path of this function.
> >>>>>>>>
> >>>>>>>>>
> >>>>>>>>> More comments to come tomorrow.  I'm especially worried about this API racing
> >>>>>>>>> with a remote processor being removed or detached.
> >>>>>>>>>
> >>>>>>>>
> >>>>>>>> Looking at what is done in wkup_m3_ipc_probe(), it should be possible to call
> >>>>>>>> rproc_get_by_handle() here and that would make sure the remote processor doesn't
> >>>>>>>> go away before the end of the function.
> >>>>>>>>
> >>>>>>>> More comments to come...
> >>>>>>
> >>>>>> It is possible to call rproc_get_by_handle() here instead of
> >>>>>> __pru_get_proc(), but that would not provide multiple functionality.
> >>>>>>
> >>>>>> The API rproc_get_by_handle() returns rproc handle on success, and NULL on
> >>>>>> failure where as __pru_get_proc() returns ERR_PTR on failure which provides
> >>>>>> multiple functionality and opportunity for us to distinguish between
> >>>>>> multiple errors.
> >>>>>>
> >>>>>> So we have these three options.
> >>>>>>
> >>>>>> 1. If we're using the API rproc_get_by_handle() and we want the multiple
> >>>>>> ERR_PTR on failure then we will need to change the API rproc_get_by_handle()
> >>>>>> and also all the functions that uses rproc_get_by_handle().
> >>>>>>
> >>>>>
> >>>>> Not optimal.
> >>>>>
> >>>>>> 2. Keep the API rproc_get_by_handle() as it is. That will restrict us from
> >>>>>> using multiple ERR_PTR on different kinds of error.
> >>>>>>
> >>>>>
> >>>>> Not optimal.
> >>>>>
> >>>>>> 3. Instead of using rproc_get_by_handle(), keep using __pru_get_proc(). This
> >>>>>> will make sure we have the proper ERR_PTR to retrun for different kinds of
> >>>>>> errors.
> >>>>>>
> >>>>>
> >>>>> Unacceptable for the reason I already stated.
> >>>>>
> >>>>>> Please let me know which option to continue with.
> >>>>>
> >>>>> I suggest building a wrapper that does everything you want around rproc_get_by_phandle().
> >>
> >> We can introduce a new API __rproc_get_by_phandle() similar to the API
> >> rproc_get_by_phandle(). The new API __rproc_get_by_phandle() will do all the
> >> functionality of getting the rproc. On success it will return rproc and on
> >> failure it will return the different ERR_PTR.
> >> If rproc is not probed yet, it will return ERR_PTR(-EPROBE_DEFER).
> >>
> >> This will make sure that we're getting different error codes for different
> >> errors from pru_rproc_get().
> >>
> >> The old API rproc_get_by_handle() will invoke the new API. On success the new
> >> API will return rproc and the old API will also return rproc. On failure the
> >> new API will return different error codes while the old API will preserve it's
> >> nature and return NULL.
> > 
> > I meant to create a wrapper around rproc_get_by_handle() that is local
> > to pru_rproc.c.  That way you can enact the behavior you want without
> > having to constrain others in this specific design.
> > 
> > Thanks,
> > Mathieu
> > 
> As The API rproc_get_by_phandle() returns null on failure, I am not able to
> create a wrapper around this API, which is local to pru_rproc.c to enact the
> different behaviors that I want. As the API is only returning null on failure
> but pru_rproc_get needs different error codes for different failures. We can
> create a wrapper but that won't be able to provide different error codes that
> we want in pru_rproc_get().

I took a closer look at __pru_rproc_get()...

Calling of_property_read_u32_index() instead of of_parse_phandle() should do the
primary sanetary check.  After that call rproc_get_by_phandle() and return
EPROBE_DEFER if it returns NULL.  If not then you have an rproc with a platform
device that you can check with is_pru_rproc() and return whatever error code you
want if it isn't to your liking.  

> 
> This is why I am suggesting to introduce new API [__rproc_get_by_phandle()]
> similar to rproc_get_by_handle(). The new API can return different error codes
> on different error. While the older API [rproc_get_by_handle()] will call the
> new API and on failure on new API the older API will just return null
> irrespective of the error code returned by the new API. Thus preserving it's
> original design.
> 
> This way we can enact the behavior that we want by calling the new API while
> the other APIs dependent on the older API, can still call
> rproc_get_by_phandle() without any change to the existing design.
> 
> We can have a little discussion to discuss this further on some IRC channel or
> meeting if it's possible.
> 
> Thanks,
> Danish
> 
> 
> >>
> >> The API __rproc_get_by_handle will look like this.
> >>
> >> struct rproc *__rproc_get_by_handle(phandle phandle)
> >> {
> >>         struct rproc *rproc = NULL, *r;
> >>         struct device_node *np;
> >>
> >>         np = of_find_node_by_phandle(phandle);
> >>         if (!np)
> >>                 return NULL;
> >>
> >>         rcu_read_lock();
> >>         list_for_each_entry_rcu(r, &rproc_list, node) {
> >>                 if (r->dev.parent && r->dev.parent->of_node == np) {
> >>                         /*prevent underlying implementation from being removed */
> >>                         if (!try_module_get(r->dev.parent->driver->owner)) {
> >>                                 dev_err(&r->dev, "can't get owner\n");
> >>                                 break;
> >>                         }
> >>
> >>                         rproc = r;
> >>                         get_device(&rproc->dev);
> >>                         break;
> >>                 }
> >>         }
> >>         rcu_read_unlock();
> >>
> >>         of_node_put(np);
> >>
> >>         if(!rproc)
> >>                 return ERR_PTR(-EPROBE_DEFER);
> >>
> >>         return rproc;
> >> }
> >>
> >> The API rproc_get_by_handle() will look like this.
> >>
> >> struct rproc *rproc_get_by_phandle(phandle phandle)
> >> {
> >>         struct rproc *rproc = NULL;
> >>
> >>         rproc = __rproc_get_by_handle(phandle);
> >>
> >>         if(!rproc || IS_ERR(rproc))
> >>                 return NULL;
> >>
> >>         return rproc;
> >> }
> >>
> >> This way in pru_rproc_get(), we'll get the rproc by phandle and we'll still
> >> return different error codes depending upon failure cases. We'll also be able
> >> to preserve the actual functionality of rproc_get_by_phandle() so that the
> >> other APIs using rproc_get_by_phandle() won't get affected.
> >>
> >> Please let me know if this looks good.
> >>
> >>>>>
> >>>>
> >>>> Sure, I'll do that. I'll share this change in v6 of this patch series.
> >>>
> >>> I'm able to test the TI prueth driver from the ti-linux-5.10.y tree [1] on a
> >>> AM5749 cpu (custom board). But I need a more recent kernel (at least 5.15) to
> >>> support other devices recently added to the Linux kernel (wifi6 module and an
> >>> ethernet switch). Also it would be nice if this driver is finally merged in the
> >>> Linux kernel.
> >>>
> >>> Maybe I can help to test this series but I noticed it only provide the driver
> >>> for TI AM654x cpus [2]. Can you also provide patches for basic EMAC support with
> >>> the TI AM574x too? (I don't need advanced features like frame timestamping, HSR
> >>> etc).
> >>>
> >>> Also, what about patches present in the ti-linux-kernel tree and not included
> >>> this this series? Especially patches that modify the kernel network stack [3]
> >>> (net/rpmsg: add support for new rpmsg sockets). Is this new socket protocol
> >>> really needed?
> >>>
> >>> Notice the patch adding the rpmsg sockets [3] already conflict with the upstream
> >>> kernel since the AF_MCTP definition now use the value temporarly used by
> >>> AF_RPMSG [4].
> >>>
> >>> Can you send an updated version of the complete series?
> >>>
> >>> Thanks!
> >>>
> >>> [1] https://git.ti.com/cgit/ti-linux-kernel/ti-linux-kernel/log/?h=ti-linux-5.10.y
> >>>
> >>> [2] https://lore.kernel.org/linux-remoteproc/20220406094358.7895-1-p-mohan@ti.com/
> >>>
> >>> [3]
> >>> https://git.ti.com/cgit/ti-linux-kernel/ti-linux-kernel/commit/?h=ti-linux-5.10.y&id=f4b978a978c38149f712ddd137f12ed5fb914161
> >>>
> >>> [4]
> >>> https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=bc49d8169aa72295104f1558830c568efb946315
> >>>
> >>> Best regards,
> >>> Romain
> >>>
> >>>
> >>>>
> >>>>>>
> >>>>>> Thanks,
> >>>>>> Danish
> >>>>>>
> >>>>>>
> >>>>>>>>
> >>>>>>>>
> >>>>>>>>> Thanks,
> >>>>>>>>> Mathieu
> >>>>>>>>>
> >>>>>>>>>> +
> >>>>>>>>>> +      rproc = __pru_rproc_get(np, index);
> >>>>>>>>>> +      if (IS_ERR(rproc))
> >>>>>>>>>> +              return rproc;
> >>>>>>>>>> +
> >>>>>>>>>> +      pru = rproc->priv;
> >>>>>>>>>> +      dev = &rproc->dev;
> >>>>>>>>>> +
> >>>>>>>>>> +      mutex_lock(&pru->lock);
> >>>>>>>>>> +
> >>>>>>>>>> +      if (pru->client_np) {
> >>>>>>>>>> +              mutex_unlock(&pru->lock);
> >>>>>>>>>> +              put_device(dev);
> >>>>>>>>>> +              return ERR_PTR(-EBUSY);
> >>>>>>>>>> +      }
> >>>>>>>>>> +
> >>>>>>>>>> +      pru->client_np = np;
> >>>>>>>>>> +
> >>>>>>>>>> +      mutex_unlock(&pru->lock);
> >>>>>>>>>> +
> >>>>>>>>>> +      if (pru_id)
> >>>>>>>>>> +              *pru_id = pru->id;
> >>>>>>>>>> +
> >>>>>>>>>> +      return rproc;
> >>>>>>>>>> +}
> >>>>>>>>>> +EXPORT_SYMBOL_GPL(pru_rproc_get);
> >>>>>>>>>> +
> >>>>>>>>>> +/**
> >>>>>>>>>> + * pru_rproc_put() - release the PRU rproc resource
> >>>>>>>>>> + * @rproc: the rproc resource to release
> >>>>>>>>>> + *
> >>>>>>>>>> + * Releases the PRU rproc resource and makes it available to other
> >>>>>>>>>> + * users.
> >>>>>>>>>> + */
> >>>>>>>>>> +void pru_rproc_put(struct rproc *rproc)
> >>>>>>>>>> +{
> >>>>>>>>>> +      struct pru_rproc *pru;
> >>>>>>>>>> +
> >>>>>>>>>> +      if (IS_ERR_OR_NULL(rproc) || !is_pru_rproc(rproc->dev.parent))
> >>>>>>>>>> +              return;
> >>>>>>>>>> +
> >>>>>>>>>> +      pru = rproc->priv;
> >>>>>>>>>> +
> >>>>>>>>>> +      mutex_lock(&pru->lock);
> >>>>>>>>>> +
> >>>>>>>>>> +      if (!pru->client_np) {
> >>>>>>>>>> +              mutex_unlock(&pru->lock);
> >>>>>>>>>> +              return;
> >>>>>>>>>> +      }
> >>>>>>>>>> +
> >>>>>>>>>> +      pru->client_np = NULL;
> >>>>>>>>>> +      mutex_unlock(&pru->lock);
> >>>>>>>>>> +
> >>>>>>>>>> +      put_device(&rproc->dev);
> >>>>>>>>>> +}
> >>>>>>>>>> +EXPORT_SYMBOL_GPL(pru_rproc_put);
> >>>>>>>>>> +
> >>>>>>>>>>   static inline u32 pru_debug_read_reg(struct pru_rproc *pru, unsigned int reg)
> >>>>>>>>>>   {
> >>>>>>>>>>        return readl_relaxed(pru->mem_regions[PRU_IOMEM_DEBUG].va + reg);
> >>>>>>>>>> @@ -438,7 +563,7 @@ static void *pru_d_da_to_va(struct pru_rproc *pru, u32 da, size_t len)
> >>>>>>>>>>        dram0 = pruss->mem_regions[PRUSS_MEM_DRAM0];
> >>>>>>>>>>        dram1 = pruss->mem_regions[PRUSS_MEM_DRAM1];
> >>>>>>>>>>        /* PRU1 has its local RAM addresses reversed */
> >>>>>>>>>> -      if (pru->id == 1)
> >>>>>>>>>> +      if (pru->id == PRUSS_PRU1)
> >>>>>>>>>>                swap(dram0, dram1);
> >>>>>>>>>>        shrd_ram = pruss->mem_regions[PRUSS_MEM_SHRD_RAM2];
> >>>>>>>>>> @@ -747,14 +872,14 @@ static int pru_rproc_set_id(struct pru_rproc *pru)
> >>>>>>>>>>        case RTU0_IRAM_ADDR_MASK:
> >>>>>>>>>>                fallthrough;
> >>>>>>>>>>        case PRU0_IRAM_ADDR_MASK:
> >>>>>>>>>> -              pru->id = 0;
> >>>>>>>>>> +              pru->id = PRUSS_PRU0;
> >>>>>>>>>>                break;
> >>>>>>>>>>        case TX_PRU1_IRAM_ADDR_MASK:
> >>>>>>>>>>                fallthrough;
> >>>>>>>>>>        case RTU1_IRAM_ADDR_MASK:
> >>>>>>>>>>                fallthrough;
> >>>>>>>>>>        case PRU1_IRAM_ADDR_MASK:
> >>>>>>>>>> -              pru->id = 1;
> >>>>>>>>>> +              pru->id = PRUSS_PRU1;
> >>>>>>>>>>                break;
> >>>>>>>>>>        default:
> >>>>>>>>>>                ret = -EINVAL;
> >>>>>>>>>> @@ -816,6 +941,8 @@ static int pru_rproc_probe(struct platform_device *pdev)
> >>>>>>>>>>        pru->pruss = platform_get_drvdata(ppdev);
> >>>>>>>>>>        pru->rproc = rproc;
> >>>>>>>>>>        pru->fw_name = fw_name;
> >>>>>>>>>> +      pru->client_np = NULL;
> >>>>>>>>>> +      mutex_init(&pru->lock);
> >>>>>>>>>>        for (i = 0; i < ARRAY_SIZE(mem_names); i++) {
> >>>>>>>>>>                res = platform_get_resource_byname(pdev, IORESOURCE_MEM,
> >>>>>>>>>> @@ -903,7 +1030,7 @@ MODULE_DEVICE_TABLE(of, pru_rproc_match);
> >>>>>>>>>>   static struct platform_driver pru_rproc_driver = {
> >>>>>>>>>>        .driver = {
> >>>>>>>>>> -              .name   = "pru-rproc",
> >>>>>>>>>> +              .name   = PRU_RPROC_DRVNAME,
> >>>>>>>>>>                .of_match_table = pru_rproc_match,
> >>>>>>>>>>                .suppress_bind_attrs = true,
> >>>>>>>>>>        },
> >>>>>>>>>> @@ -915,5 +1042,6 @@ module_platform_driver(pru_rproc_driver);
> >>>>>>>>>>   MODULE_AUTHOR("Suman Anna <s-anna@ti.com>");
> >>>>>>>>>>   MODULE_AUTHOR("Andrew F. Davis <afd@ti.com>");
> >>>>>>>>>>   MODULE_AUTHOR("Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>");
> >>>>>>>>>> +MODULE_AUTHOR("Puranjay Mohan <p-mohan@ti.com>");
> >>>>>>>>>>   MODULE_DESCRIPTION("PRU-ICSS Remote Processor Driver");
> >>>>>>>>>>   MODULE_LICENSE("GPL v2");
> >>>>>>>>>> diff --git a/include/linux/pruss.h b/include/linux/pruss.h
> >>>>>>>>>> new file mode 100644
> >>>>>>>>>> index 000000000000..fdc719b43db0
> >>>>>>>>>> --- /dev/null
> >>>>>>>>>> +++ b/include/linux/pruss.h
> >>>>>>>>>> @@ -0,0 +1,56 @@
> >>>>>>>>>> +/* SPDX-License-Identifier: GPL-2.0-only */
> >>>>>>>>>> +/**
> >>>>>>>>>> + * PRU-ICSS Subsystem user interfaces
> >>>>>>>>>> + *
> >>>>>>>>>> + * Copyright (C) 2015-2022 Texas Instruments Incorporated - http://www.ti.com
> >>>>>>>>>> + *    Suman Anna <s-anna@ti.com>
> >>>>>>>>>> + */
> >>>>>>>>>> +
> >>>>>>>>>> +#ifndef __LINUX_PRUSS_H
> >>>>>>>>>> +#define __LINUX_PRUSS_H
> >>>>>>>>>> +
> >>>>>>>>>> +#include <linux/device.h>
> >>>>>>>>>> +#include <linux/types.h>
> >>>>>>>>>> +
> >>>>>>>>>> +#define PRU_RPROC_DRVNAME "pru-rproc"
> >>>>>>>>>> +
> >>>>>>>>>> +/*
> >>>>>>>>>> + * enum pruss_pru_id - PRU core identifiers
> >>>>>>>>>> + */
> >>>>>>>>>> +enum pruss_pru_id {
> >>>>>>>>>> +      PRUSS_PRU0 = 0,
> >>>>>>>>>> +      PRUSS_PRU1,
> >>>>>>>>>> +      PRUSS_NUM_PRUS,
> >>>>>>>>>> +};
> >>>>>>>>>> +
> >>>>>>>>>> +struct device_node;
> >>>>>>>>>> +
> >>>>>>>>>> +#if IS_ENABLED(CONFIG_PRU_REMOTEPROC)
> >>>>>>>>>> +
> >>>>>>>>>> +struct rproc *pru_rproc_get(struct device_node *np, int index,
> >>>>>>>>>> +                          enum pruss_pru_id *pru_id);
> >>>>>>>>>> +void pru_rproc_put(struct rproc *rproc);
> >>>>>>>>>> +
> >>>>>>>>>> +#else
> >>>>>>>>>> +
> >>>>>>>>>> +static inline struct rproc *
> >>>>>>>>>> +pru_rproc_get(struct device_node *np, int index, enum pruss_pru_id *pru_id)
> >>>>>>>>>> +{
> >>>>>>>>>> +      return ERR_PTR(-EOPNOTSUPP);
> >>>>>>>>>> +}
> >>>>>>>>>> +
> >>>>>>>>>> +static inline void pru_rproc_put(struct rproc *rproc) { }
> >>>>>>>>>> +
> >>>>>>>>>> +#endif /* CONFIG_PRU_REMOTEPROC */
> >>>>>>>>>> +
> >>>>>>>>>> +static inline bool is_pru_rproc(struct device *dev)
> >>>>>>>>>> +{
> >>>>>>>>>> +      const char *drv_name = dev_driver_string(dev);
> >>>>>>>>>> +
> >>>>>>>>>> +      if (strncmp(drv_name, PRU_RPROC_DRVNAME, sizeof(PRU_RPROC_DRVNAME)))
> >>>>>>>>>> +              return false;
> >>>>>>>>>> +
> >>>>>>>>>> +      return true;
> >>>>>>>>>> +}
> >>>>>>>>>> +
> >>>>>>>>>> +#endif /* __LINUX_PRUSS_H */
> >>>>>>>>>> --
> >>>>>>>>>> 2.17.1
> >>>>>>>>>>
> >>>>
> >>>

[PATCH v5 3/6] remoteproc: pru: Make sysfs entries read-only for PRU client driven boots

[Puranjay Mohan]

From: Suman Anna <s-anna@ti.com>

The PRU remoteproc driver is not configured for 'auto-boot' by default,
and allows to be booted either by in-kernel PRU client drivers or by
userspace using the generic remoteproc sysfs interfaces. The sysfs
interfaces should not be permitted to change the remoteproc firmwares
or states when a PRU is being managed by an in-kernel client driver.
Use the newly introduced remoteproc generic 'sysfs_read_only' flag to
provide these restrictions by setting and clearing it appropriately
during the PRU acquire and release steps.

Signed-off-by: Suman Anna <s-anna@ti.com>
Co-developed-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
Signed-off-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
---
 drivers/remoteproc/pru_rproc.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
index 7a35b400287a..9fed3e0372d3 100644
--- a/drivers/remoteproc/pru_rproc.c
+++ b/drivers/remoteproc/pru_rproc.c
@@ -231,6 +231,7 @@ struct rproc *pru_rproc_get(struct device_node *np, int index,
 	}
 
 	pru->client_np = np;
+	rproc->sysfs_read_only = true;
 
 	mutex_unlock(&pru->lock);
 
@@ -265,6 +266,7 @@ void pru_rproc_put(struct rproc *rproc)
 	}
 
 	pru->client_np = NULL;
+	rproc->sysfs_read_only = false;
 	mutex_unlock(&pru->lock);
 
 	put_device(&rproc->dev);
-- 
2.17.1

[PATCH v5 4/6] remoteproc: pru: Add pru_rproc_set_ctable() function

[Puranjay Mohan]

From: Roger Quadros <rogerq@ti.com>

Some firmwares expect the OS drivers to configure the CTABLE
entries publishing dynamically allocated memory regions. For
example, the PRU Ethernet firmwares use the C28 and C30 entries
for retrieving the Shared RAM and System SRAM (OCMC) areas
allocated by the PRU Ethernet client driver.

Provide a way for users to do that through a new API,
pru_rproc_set_ctable(). The API returns 0 on success and
a negative value on error.

NOTE:
The programmable CTABLE entries are typically re-programmed by
the PRU firmwares when dealing with a certain block of memory
during block processing. This API provides an interface to the
PRU client drivers to publish a dynamically allocated memory
block with the PRU firmware using a CTABLE entry instead of a
negotiated address in shared memory. Additional synchronization
may be needed between the PRU client drivers and firmwares if
different addresses needs to be published at run-time reusing
the same CTABLE entry.

Co-developed-by: Andrew F. Davis <afd@ti.com>
Signed-off-by: Andrew F. Davis <afd@ti.com>
Co-developed-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Roger Quadros <rogerq@ti.com>
Co-developed-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
Signed-off-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
---
 drivers/remoteproc/pru_rproc.c | 59 ++++++++++++++++++++++++++++++++++
 include/linux/pruss.h          | 22 +++++++++++++
 2 files changed, 81 insertions(+)

diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
index 9fed3e0372d3..d06b763e995e 100644
--- a/drivers/remoteproc/pru_rproc.c
+++ b/drivers/remoteproc/pru_rproc.c
@@ -119,6 +119,7 @@ struct pru_private_data {
  * @mapped_irq: virtual interrupt numbers of created fw specific mapping
  * @pru_interrupt_map: pointer to interrupt mapping description (firmware)
  * @pru_interrupt_map_sz: pru_interrupt_map size
+ * @rmw_lock: lock for read, modify, write operations on registers
  * @dbg_single_step: debug state variable to set PRU into single step mode
  * @dbg_continuous: debug state variable to restore PRU execution mode
  * @evt_count: number of mapped events
@@ -136,6 +137,7 @@ struct pru_rproc {
 	unsigned int *mapped_irq;
 	struct pru_irq_rsc *pru_interrupt_map;
 	size_t pru_interrupt_map_sz;
+	spinlock_t rmw_lock; /* register access lock */
 	u32 dbg_single_step;
 	u32 dbg_continuous;
 	u8 evt_count;
@@ -152,6 +154,23 @@ void pru_control_write_reg(struct pru_rproc *pru, unsigned int reg, u32 val)
 	writel_relaxed(val, pru->mem_regions[PRU_IOMEM_CTRL].va + reg);
 }
 
+static inline
+void pru_control_set_reg(struct pru_rproc *pru, unsigned int reg,
+			 u32 mask, u32 set)
+{
+	u32 val;
+	unsigned long flags;
+
+	spin_lock_irqsave(&pru->rmw_lock, flags);
+
+	val = pru_control_read_reg(pru, reg);
+	val &= ~mask;
+	val |= (set & mask);
+	pru_control_write_reg(pru, reg, val);
+
+	spin_unlock_irqrestore(&pru->rmw_lock, flags);
+}
+
 static struct rproc *__pru_rproc_get(struct device_node *np, int index)
 {
 	struct device_node *rproc_np = NULL;
@@ -273,6 +292,45 @@ void pru_rproc_put(struct rproc *rproc)
 }
 EXPORT_SYMBOL_GPL(pru_rproc_put);
 
+/**
+ * pru_rproc_set_ctable() - set the constant table index for the PRU
+ * @rproc: the rproc instance of the PRU
+ * @c: constant table index to set
+ * @addr: physical address to set it to
+ *
+ * Return: 0 on success, or errno in error case.
+ */
+int pru_rproc_set_ctable(struct rproc *rproc, enum pru_ctable_idx c, u32 addr)
+{
+	struct pru_rproc *pru = rproc->priv;
+	unsigned int reg;
+	u32 mask, set;
+	u16 idx;
+	u16 idx_mask;
+
+	if (IS_ERR_OR_NULL(rproc))
+		return -EINVAL;
+
+	if (!rproc->dev.parent || !is_pru_rproc(rproc->dev.parent))
+		return -ENODEV;
+
+	/* pointer is 16 bit and index is 8-bit so mask out the rest */
+	idx_mask = (c >= PRU_C28) ? 0xFFFF : 0xFF;
+
+	/* ctable uses bit 8 and upwards only */
+	idx = (addr >> 8) & idx_mask;
+
+	/* configurable ctable (i.e. C24) starts at PRU_CTRL_CTBIR0 */
+	reg = PRU_CTRL_CTBIR0 + 4 * (c >> 1);
+	mask = idx_mask << (16 * (c & 1));
+	set = idx << (16 * (c & 1));
+
+	pru_control_set_reg(pru, reg, mask, set);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(pru_rproc_set_ctable);
+
 static inline u32 pru_debug_read_reg(struct pru_rproc *pru, unsigned int reg)
 {
 	return readl_relaxed(pru->mem_regions[PRU_IOMEM_DEBUG].va + reg);
@@ -944,6 +1002,7 @@ static int pru_rproc_probe(struct platform_device *pdev)
 	pru->rproc = rproc;
 	pru->fw_name = fw_name;
 	pru->client_np = NULL;
+	spin_lock_init(&pru->rmw_lock);
 	mutex_init(&pru->lock);
 
 	for (i = 0; i < ARRAY_SIZE(mem_names); i++) {
diff --git a/include/linux/pruss.h b/include/linux/pruss.h
index fdc719b43db0..d830e20056c7 100644
--- a/include/linux/pruss.h
+++ b/include/linux/pruss.h
@@ -23,13 +23,29 @@ enum pruss_pru_id {
 	PRUSS_NUM_PRUS,
 };
 
+/*
+ * enum pru_ctable_idx - Configurable Constant table index identifiers
+ */
+enum pru_ctable_idx {
+	PRU_C24 = 0,
+	PRU_C25,
+	PRU_C26,
+	PRU_C27,
+	PRU_C28,
+	PRU_C29,
+	PRU_C30,
+	PRU_C31,
+};
+
 struct device_node;
+struct rproc;
 
 #if IS_ENABLED(CONFIG_PRU_REMOTEPROC)
 
 struct rproc *pru_rproc_get(struct device_node *np, int index,
 			    enum pruss_pru_id *pru_id);
 void pru_rproc_put(struct rproc *rproc);
+int pru_rproc_set_ctable(struct rproc *rproc, enum pru_ctable_idx c, u32 addr);
 
 #else
 
@@ -41,6 +57,12 @@ pru_rproc_get(struct device_node *np, int index, enum pruss_pru_id *pru_id)
 
 static inline void pru_rproc_put(struct rproc *rproc) { }
 
+static inline int pru_rproc_set_ctable(struct rproc *rproc,
+				       enum pru_ctable_idx c, u32 addr)
+{
+	return -EOPNOTSUPP;
+}
+
 #endif /* CONFIG_PRU_REMOTEPROC */
 
 static inline bool is_pru_rproc(struct device *dev)
-- 
2.17.1

Re: [PATCH v5 4/6] remoteproc: pru: Add pru_rproc_set_ctable() function

[Mathieu Poirier]

On Tue, Jun 07, 2022 at 10:26:48AM +0530, Puranjay Mohan wrote:
> From: Roger Quadros <rogerq@ti.com>
> 
> Some firmwares expect the OS drivers to configure the CTABLE
> entries publishing dynamically allocated memory regions. For
> example, the PRU Ethernet firmwares use the C28 and C30 entries
> for retrieving the Shared RAM and System SRAM (OCMC) areas
> allocated by the PRU Ethernet client driver.
> 
> Provide a way for users to do that through a new API,
> pru_rproc_set_ctable(). The API returns 0 on success and
> a negative value on error.
> 
> NOTE:
> The programmable CTABLE entries are typically re-programmed by
> the PRU firmwares when dealing with a certain block of memory
> during block processing. This API provides an interface to the
> PRU client drivers to publish a dynamically allocated memory
> block with the PRU firmware using a CTABLE entry instead of a
> negotiated address in shared memory. Additional synchronization
> may be needed between the PRU client drivers and firmwares if
> different addresses needs to be published at run-time reusing
> the same CTABLE entry.

In all this the concept of a "ctable" is not explained and as such, I have to
guess it stands for "constant table".  I also have to guess this table stores
memory addresses known to other drivers, making it possible to boot a system
without all components having access to a DT.  Again, those are all guesses
because it is not explained.  And all that guessing makes reviewing this small
patchset quite difficult.

> 
> Co-developed-by: Andrew F. Davis <afd@ti.com>
> Signed-off-by: Andrew F. Davis <afd@ti.com>
> Co-developed-by: Suman Anna <s-anna@ti.com>
> Signed-off-by: Suman Anna <s-anna@ti.com>
> Signed-off-by: Roger Quadros <rogerq@ti.com>
> Co-developed-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
> Signed-off-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
> Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
> ---
>  drivers/remoteproc/pru_rproc.c | 59 ++++++++++++++++++++++++++++++++++
>  include/linux/pruss.h          | 22 +++++++++++++
>  2 files changed, 81 insertions(+)
> 
> diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
> index 9fed3e0372d3..d06b763e995e 100644
> --- a/drivers/remoteproc/pru_rproc.c
> +++ b/drivers/remoteproc/pru_rproc.c
> @@ -119,6 +119,7 @@ struct pru_private_data {
>   * @mapped_irq: virtual interrupt numbers of created fw specific mapping
>   * @pru_interrupt_map: pointer to interrupt mapping description (firmware)
>   * @pru_interrupt_map_sz: pru_interrupt_map size
> + * @rmw_lock: lock for read, modify, write operations on registers
>   * @dbg_single_step: debug state variable to set PRU into single step mode
>   * @dbg_continuous: debug state variable to restore PRU execution mode
>   * @evt_count: number of mapped events
> @@ -136,6 +137,7 @@ struct pru_rproc {
>  	unsigned int *mapped_irq;
>  	struct pru_irq_rsc *pru_interrupt_map;
>  	size_t pru_interrupt_map_sz;
> +	spinlock_t rmw_lock; /* register access lock */
>  	u32 dbg_single_step;
>  	u32 dbg_continuous;
>  	u8 evt_count;
> @@ -152,6 +154,23 @@ void pru_control_write_reg(struct pru_rproc *pru, unsigned int reg, u32 val)
>  	writel_relaxed(val, pru->mem_regions[PRU_IOMEM_CTRL].va + reg);
>  }
>  
> +static inline
> +void pru_control_set_reg(struct pru_rproc *pru, unsigned int reg,
> +			 u32 mask, u32 set)
> +{
> +	u32 val;
> +	unsigned long flags;
> +
> +	spin_lock_irqsave(&pru->rmw_lock, flags);
> +
> +	val = pru_control_read_reg(pru, reg);
> +	val &= ~mask;
> +	val |= (set & mask);
> +	pru_control_write_reg(pru, reg, val);
> +
> +	spin_unlock_irqrestore(&pru->rmw_lock, flags);
> +}
> +
>  static struct rproc *__pru_rproc_get(struct device_node *np, int index)
>  {
>  	struct device_node *rproc_np = NULL;
> @@ -273,6 +292,45 @@ void pru_rproc_put(struct rproc *rproc)
>  }
>  EXPORT_SYMBOL_GPL(pru_rproc_put);
>  
> +/**
> + * pru_rproc_set_ctable() - set the constant table index for the PRU
> + * @rproc: the rproc instance of the PRU
> + * @c: constant table index to set
> + * @addr: physical address to set it to
> + *
> + * Return: 0 on success, or errno in error case.
> + */
> +int pru_rproc_set_ctable(struct rproc *rproc, enum pru_ctable_idx c, u32 addr)
> +{
> +	struct pru_rproc *pru = rproc->priv;
> +	unsigned int reg;
> +	u32 mask, set;
> +	u16 idx;
> +	u16 idx_mask;
> +
> +	if (IS_ERR_OR_NULL(rproc))
> +		return -EINVAL;
> +
> +	if (!rproc->dev.parent || !is_pru_rproc(rproc->dev.parent))
> +		return -ENODEV;
> +
> +	/* pointer is 16 bit and index is 8-bit so mask out the rest */
> +	idx_mask = (c >= PRU_C28) ? 0xFFFF : 0xFF;
> +
> +	/* ctable uses bit 8 and upwards only */
> +	idx = (addr >> 8) & idx_mask;
> +
> +	/* configurable ctable (i.e. C24) starts at PRU_CTRL_CTBIR0 */
> +	reg = PRU_CTRL_CTBIR0 + 4 * (c >> 1);
> +	mask = idx_mask << (16 * (c & 1));
> +	set = idx << (16 * (c & 1));
> +

Please add comments that describe the content and format of the CTABLE.
Otherwise this code becomes unmaintainable.

> +	pru_control_set_reg(pru, reg, mask, set);
> +
> +	return 0;
> +}
> +EXPORT_SYMBOL_GPL(pru_rproc_set_ctable);
> +
>  static inline u32 pru_debug_read_reg(struct pru_rproc *pru, unsigned int reg)
>  {
>  	return readl_relaxed(pru->mem_regions[PRU_IOMEM_DEBUG].va + reg);
> @@ -944,6 +1002,7 @@ static int pru_rproc_probe(struct platform_device *pdev)
>  	pru->rproc = rproc;
>  	pru->fw_name = fw_name;
>  	pru->client_np = NULL;
> +	spin_lock_init(&pru->rmw_lock);
>  	mutex_init(&pru->lock);
>  
>  	for (i = 0; i < ARRAY_SIZE(mem_names); i++) {
> diff --git a/include/linux/pruss.h b/include/linux/pruss.h
> index fdc719b43db0..d830e20056c7 100644
> --- a/include/linux/pruss.h
> +++ b/include/linux/pruss.h
> @@ -23,13 +23,29 @@ enum pruss_pru_id {
>  	PRUSS_NUM_PRUS,
>  };
>  
> +/*
> + * enum pru_ctable_idx - Configurable Constant table index identifiers
> + */
> +enum pru_ctable_idx {
> +	PRU_C24 = 0,
> +	PRU_C25,
> +	PRU_C26,
> +	PRU_C27,
> +	PRU_C28,
> +	PRU_C29,
> +	PRU_C30,
> +	PRU_C31,
> +};
> +
>  struct device_node;
> +struct rproc;
>  
>  #if IS_ENABLED(CONFIG_PRU_REMOTEPROC)
>  
>  struct rproc *pru_rproc_get(struct device_node *np, int index,
>  			    enum pruss_pru_id *pru_id);
>  void pru_rproc_put(struct rproc *rproc);
> +int pru_rproc_set_ctable(struct rproc *rproc, enum pru_ctable_idx c, u32 addr);
>  
>  #else
>  
> @@ -41,6 +57,12 @@ pru_rproc_get(struct device_node *np, int index, enum pruss_pru_id *pru_id)
>  
>  static inline void pru_rproc_put(struct rproc *rproc) { }
>  
> +static inline int pru_rproc_set_ctable(struct rproc *rproc,
> +				       enum pru_ctable_idx c, u32 addr)
> +{
> +	return -EOPNOTSUPP;
> +}
> +
>  #endif /* CONFIG_PRU_REMOTEPROC */
>  
>  static inline bool is_pru_rproc(struct device *dev)
> -- 
> 2.17.1
> 

[PATCH v5 5/6] remoteproc: pru: Configure firmware based on client setup

[Puranjay Mohan]

From: Tero Kristo <t-kristo@ti.com>

Client device node property firmware-name is now used to configure
firmware for the PRU instances. The default firmware is also
restored once releasing the PRU resource.

Co-developed-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Tero Kristo <t-kristo@ti.com>
Co-developed-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
Signed-off-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
---
 drivers/remoteproc/pru_rproc.c | 35 ++++++++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
index d06b763e995e..2977eb50631b 100644
--- a/drivers/remoteproc/pru_rproc.c
+++ b/drivers/remoteproc/pru_rproc.c
@@ -171,6 +171,23 @@ void pru_control_set_reg(struct pru_rproc *pru, unsigned int reg,
 	spin_unlock_irqrestore(&pru->rmw_lock, flags);
 }
 
+/**
+ * pru_rproc_set_firmware() - set firmware for a pru core
+ * @rproc: the rproc instance of the PRU
+ * @fw_name: the new firmware name, or NULL if default is desired
+ *
+ * Return: 0 on success, or errno in error case.
+ */
+static int pru_rproc_set_firmware(struct rproc *rproc, const char *fw_name)
+{
+	struct pru_rproc *pru = rproc->priv;
+
+	if (!fw_name)
+		fw_name = pru->fw_name;
+
+	return rproc_set_firmware(rproc, fw_name);
+}
+
 static struct rproc *__pru_rproc_get(struct device_node *np, int index)
 {
 	struct device_node *rproc_np = NULL;
@@ -231,6 +248,8 @@ struct rproc *pru_rproc_get(struct device_node *np, int index,
 	struct rproc *rproc;
 	struct pru_rproc *pru;
 	struct device *dev;
+	const char *fw_name;
+	int ret;
 
 	try_module_get(THIS_MODULE);
 
@@ -257,7 +276,21 @@ struct rproc *pru_rproc_get(struct device_node *np, int index,
 	if (pru_id)
 		*pru_id = pru->id;
 
+	ret = of_property_read_string_index(np, "firmware-name", index,
+					    &fw_name);
+	if (!ret) {
+		ret = pru_rproc_set_firmware(rproc, fw_name);
+		if (ret) {
+			dev_err(dev, "failed to set firmware: %d\n", ret);
+			goto err;
+		}
+	}
+
 	return rproc;
+
+err:
+	pru_rproc_put(rproc);
+	return ERR_PTR(ret);
 }
 EXPORT_SYMBOL_GPL(pru_rproc_get);
 
@@ -277,6 +310,8 @@ void pru_rproc_put(struct rproc *rproc)
 
 	pru = rproc->priv;
 
+	pru_rproc_set_firmware(rproc, NULL);
+
 	mutex_lock(&pru->lock);
 
 	if (!pru->client_np) {
-- 
2.17.1

[PATCH v5 6/6] remoteproc: pru: add support for configuring GPMUX based on client setup

[Puranjay Mohan]

From: Tero Kristo <t-kristo@ti.com>

Client device node property ti,pruss-gp-mux-sel can now be used to
configure the GPMUX config value for PRU.

Signed-off-by: Tero Kristo <t-kristo@ti.com>
[s-anna@ti.com: simplify the pru id usage]
Signed-off-by: Suman Anna <s-anna@ti.com>
Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
---
V4->v5
* This patch was included in v4 and had some checkpatch errors that have
  been resolved in v5
---
 drivers/remoteproc/pru_rproc.c | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
index 2977eb50631b..f2c6c55f0f20 100644
--- a/drivers/remoteproc/pru_rproc.c
+++ b/drivers/remoteproc/pru_rproc.c
@@ -123,6 +123,7 @@ struct pru_private_data {
  * @dbg_single_step: debug state variable to set PRU into single step mode
  * @dbg_continuous: debug state variable to restore PRU execution mode
  * @evt_count: number of mapped events
+ * @gpmux_save: saved value for gpmux config
  */
 struct pru_rproc {
 	int id;
@@ -141,6 +142,7 @@ struct pru_rproc {
 	u32 dbg_single_step;
 	u32 dbg_continuous;
 	u8 evt_count;
+	u8 gpmux_save;
 };
 
 static inline u32 pru_control_read_reg(struct pru_rproc *pru, unsigned int reg)
@@ -250,6 +252,7 @@ struct rproc *pru_rproc_get(struct device_node *np, int index,
 	struct device *dev;
 	const char *fw_name;
 	int ret;
+	u32 mux;
 
 	try_module_get(THIS_MODULE);
 
@@ -273,6 +276,22 @@ struct rproc *pru_rproc_get(struct device_node *np, int index,
 
 	mutex_unlock(&pru->lock);
 
+	ret = pruss_cfg_get_gpmux(pru->pruss, pru->id, &pru->gpmux_save);
+	if (ret) {
+		dev_err(dev, "failed to get cfg gpmux: %d\n", ret);
+		goto err;
+	}
+
+	ret = of_property_read_u32_index(np, "ti,pruss-gp-mux-sel", index,
+					 &mux);
+	if (!ret) {
+		ret = pruss_cfg_set_gpmux(pru->pruss, pru->id, mux);
+		if (ret) {
+			dev_err(dev, "failed to set cfg gpmux: %d\n", ret);
+			goto err;
+		}
+	}
+
 	if (pru_id)
 		*pru_id = pru->id;
 
@@ -310,6 +329,7 @@ void pru_rproc_put(struct rproc *rproc)
 
 	pru = rproc->priv;
 
+	pruss_cfg_set_gpmux(pru->pruss, pru->id, pru->gpmux_save);
 	pru_rproc_set_firmware(rproc, NULL);
 
 	mutex_lock(&pru->lock);
-- 
2.17.1

Re: [PATCH v5 6/6] remoteproc: pru: add support for configuring GPMUX based on client setup

[Mathieu Poirier]

On Tue, Jun 07, 2022 at 10:26:50AM +0530, Puranjay Mohan wrote:
> From: Tero Kristo <t-kristo@ti.com>
> 
> Client device node property ti,pruss-gp-mux-sel can now be used to
> configure the GPMUX config value for PRU.
> 
> Signed-off-by: Tero Kristo <t-kristo@ti.com>
> [s-anna@ti.com: simplify the pru id usage]
> Signed-off-by: Suman Anna <s-anna@ti.com>
> Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
> ---
> V4->v5
> * This patch was included in v4 and had some checkpatch errors that have
>   been resolved in v5
> ---
>  drivers/remoteproc/pru_rproc.c | 20 ++++++++++++++++++++
>  1 file changed, 20 insertions(+)
> 
> diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
> index 2977eb50631b..f2c6c55f0f20 100644
> --- a/drivers/remoteproc/pru_rproc.c
> +++ b/drivers/remoteproc/pru_rproc.c
> @@ -123,6 +123,7 @@ struct pru_private_data {
>   * @dbg_single_step: debug state variable to set PRU into single step mode
>   * @dbg_continuous: debug state variable to restore PRU execution mode
>   * @evt_count: number of mapped events
> + * @gpmux_save: saved value for gpmux config
>   */
>  struct pru_rproc {
>  	int id;
> @@ -141,6 +142,7 @@ struct pru_rproc {
>  	u32 dbg_single_step;
>  	u32 dbg_continuous;
>  	u8 evt_count;
> +	u8 gpmux_save;
>  };
>  
>  static inline u32 pru_control_read_reg(struct pru_rproc *pru, unsigned int reg)
> @@ -250,6 +252,7 @@ struct rproc *pru_rproc_get(struct device_node *np, int index,
>  	struct device *dev;
>  	const char *fw_name;
>  	int ret;
> +	u32 mux;
>  
>  	try_module_get(THIS_MODULE);
>  
> @@ -273,6 +276,22 @@ struct rproc *pru_rproc_get(struct device_node *np, int index,
>  
>  	mutex_unlock(&pru->lock);
>  
> +	ret = pruss_cfg_get_gpmux(pru->pruss, pru->id, &pru->gpmux_save);
> +	if (ret) {
> +		dev_err(dev, "failed to get cfg gpmux: %d\n", ret);
> +		goto err;
> +	}
> +
> +	ret = of_property_read_u32_index(np, "ti,pruss-gp-mux-sel", index,
> +					 &mux);
> +	if (!ret) {
> +		ret = pruss_cfg_set_gpmux(pru->pruss, pru->id, mux);
> +		if (ret) {
> +			dev_err(dev, "failed to set cfg gpmux: %d\n", ret);
> +			goto err;
> +		}
> +	}
> +
>  	if (pru_id)
>  		*pru_id = pru->id;
>  
> @@ -310,6 +329,7 @@ void pru_rproc_put(struct rproc *rproc)
>  
>  	pru = rproc->priv;
>  
> +	pruss_cfg_set_gpmux(pru->pruss, pru->id, pru->gpmux_save);
>  	pru_rproc_set_firmware(rproc, NULL);
>

  CC      drivers/remoteproc/pru_rproc.o
/home/mpoirier/work/remoteproc/kernel-review/drivers/remoteproc/pru_rproc.c: In function ‘pru_rproc_get’:
/home/mpoirier/work/remoteproc/kernel-review/drivers/remoteproc/pru_rproc.c:279:8: error: implicit declaration of function ‘pruss_cfg_get_gpmux’ [-Werror=implicit-function-declaration]
  279 |  ret = pruss_cfg_get_gpmux(pru->pruss, pru->id, &pru->gpmux_save);
      |        ^~~~~~~~~~~~~~~~~~~
/home/mpoirier/work/remoteproc/kernel-review/drivers/remoteproc/pru_rproc.c:288:9: error: implicit declaration of function ‘pruss_cfg_set_gpmux’ [-Werror=implicit-function-declaration]
  288 |   ret = pruss_cfg_set_gpmux(pru->pruss, pru->id, mux);
      |         ^~~~~~~~~~~~~~~~~~~

I get this on both rproc-next and today's linux next.  

>  	mutex_lock(&pru->lock);
> -- 
> 2.17.1
> 

Re: [PATCH v5 6/6] remoteproc: pru: add support for configuring GPMUX based on client setup

[Md Danish Anwar]

Hi Mathieu,

On 19/07/22 21:46, Mathieu Poirier wrote:
> On Tue, Jun 07, 2022 at 10:26:50AM +0530, Puranjay Mohan wrote:
>> From: Tero Kristo <t-kristo@ti.com>
>>
>> Client device node property ti,pruss-gp-mux-sel can now be used to
>> configure the GPMUX config value for PRU.
>>
>> Signed-off-by: Tero Kristo <t-kristo@ti.com>
>> [s-anna@ti.com: simplify the pru id usage]
>> Signed-off-by: Suman Anna <s-anna@ti.com>
>> Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
>> ---
>> V4->v5
>> * This patch was included in v4 and had some checkpatch errors that have
>>   been resolved in v5
>> ---
>>  drivers/remoteproc/pru_rproc.c | 20 ++++++++++++++++++++
>>  1 file changed, 20 insertions(+)
>>
>> diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
>> index 2977eb50631b..f2c6c55f0f20 100644
>> --- a/drivers/remoteproc/pru_rproc.c
>> +++ b/drivers/remoteproc/pru_rproc.c
>> @@ -123,6 +123,7 @@ struct pru_private_data {
>>   * @dbg_single_step: debug state variable to set PRU into single step mode
>>   * @dbg_continuous: debug state variable to restore PRU execution mode
>>   * @evt_count: number of mapped events
>> + * @gpmux_save: saved value for gpmux config
>>   */
>>  struct pru_rproc {
>>  	int id;
>> @@ -141,6 +142,7 @@ struct pru_rproc {
>>  	u32 dbg_single_step;
>>  	u32 dbg_continuous;
>>  	u8 evt_count;
>> +	u8 gpmux_save;
>>  };
>>  
>>  static inline u32 pru_control_read_reg(struct pru_rproc *pru, unsigned int reg)
>> @@ -250,6 +252,7 @@ struct rproc *pru_rproc_get(struct device_node *np, int index,
>>  	struct device *dev;
>>  	const char *fw_name;
>>  	int ret;
>> +	u32 mux;
>>  
>>  	try_module_get(THIS_MODULE);
>>  
>> @@ -273,6 +276,22 @@ struct rproc *pru_rproc_get(struct device_node *np, int index,
>>  
>>  	mutex_unlock(&pru->lock);
>>  
>> +	ret = pruss_cfg_get_gpmux(pru->pruss, pru->id, &pru->gpmux_save);
>> +	if (ret) {
>> +		dev_err(dev, "failed to get cfg gpmux: %d\n", ret);
>> +		goto err;
>> +	}
>> +
>> +	ret = of_property_read_u32_index(np, "ti,pruss-gp-mux-sel", index,
>> +					 &mux);
>> +	if (!ret) {
>> +		ret = pruss_cfg_set_gpmux(pru->pruss, pru->id, mux);
>> +		if (ret) {
>> +			dev_err(dev, "failed to set cfg gpmux: %d\n", ret);
>> +			goto err;
>> +		}
>> +	}
>> +
>>  	if (pru_id)
>>  		*pru_id = pru->id;
>>  
>> @@ -310,6 +329,7 @@ void pru_rproc_put(struct rproc *rproc)
>>  
>>  	pru = rproc->priv;
>>  
>> +	pruss_cfg_set_gpmux(pru->pruss, pru->id, pru->gpmux_save);
>>  	pru_rproc_set_firmware(rproc, NULL);
>>
> 
>   CC      drivers/remoteproc/pru_rproc.o
> /home/mpoirier/work/remoteproc/kernel-review/drivers/remoteproc/pru_rproc.c: In function ‘pru_rproc_get’:
> /home/mpoirier/work/remoteproc/kernel-review/drivers/remoteproc/pru_rproc.c:279:8: error: implicit declaration of function ‘pruss_cfg_get_gpmux’ [-Werror=implicit-function-declaration]
>   279 |  ret = pruss_cfg_get_gpmux(pru->pruss, pru->id, &pru->gpmux_save);
>       |        ^~~~~~~~~~~~~~~~~~~
> /home/mpoirier/work/remoteproc/kernel-review/drivers/remoteproc/pru_rproc.c:288:9: error: implicit declaration of function ‘pruss_cfg_set_gpmux’ [-Werror=implicit-function-declaration]
>   288 |   ret = pruss_cfg_set_gpmux(pru->pruss, pru->id, mux);
>       |         ^~~~~~~~~~~~~~~~~~~
> 
> I get this on both rproc-next and today's linux next.  

This patch is dependent on the series [2] Introduce PRU platform consumer API
https://patchwork.kernel.org/project/linux-remoteproc/cover/20220406094358.7895-1-p-mohan@ti.com/
as the api pruss_cfg_get_gpmux calls the api pruss_cfg_read and the api
pruss_cfg_set_gpmux calls the api pruss_cfg_update which are implemented by the
patch "soc: ti: pruss: Add pruss_cfg_read()/update() API" in the above series.

This error is coming as the dependent patch needs the patch "soc: ti: pruss:
Add pruss_cfg_read()/update() API" to be applied for compilation.

Thanks,
Danish.

> 
>>  	mutex_lock(&pru->lock);
>> -- 
>> 2.17.1
>>

Re: [PATCH v5 6/6] remoteproc: pru: add support for configuring GPMUX based on client setup

[Mathieu Poirier]

Good day,

On Mon, Sep 05, 2022 at 02:41:27PM +0530, Md Danish Anwar wrote:
> Hi Mathieu,
> 
> On 19/07/22 21:46, Mathieu Poirier wrote:
> > On Tue, Jun 07, 2022 at 10:26:50AM +0530, Puranjay Mohan wrote:
> >> From: Tero Kristo <t-kristo@ti.com>
> >>
> >> Client device node property ti,pruss-gp-mux-sel can now be used to
> >> configure the GPMUX config value for PRU.
> >>
> >> Signed-off-by: Tero Kristo <t-kristo@ti.com>
> >> [s-anna@ti.com: simplify the pru id usage]
> >> Signed-off-by: Suman Anna <s-anna@ti.com>
> >> Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
> >> ---
> >> V4->v5
> >> * This patch was included in v4 and had some checkpatch errors that have
> >>   been resolved in v5
> >> ---
> >>  drivers/remoteproc/pru_rproc.c | 20 ++++++++++++++++++++
> >>  1 file changed, 20 insertions(+)
> >>
> >> diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
> >> index 2977eb50631b..f2c6c55f0f20 100644
> >> --- a/drivers/remoteproc/pru_rproc.c
> >> +++ b/drivers/remoteproc/pru_rproc.c
> >> @@ -123,6 +123,7 @@ struct pru_private_data {
> >>   * @dbg_single_step: debug state variable to set PRU into single step mode
> >>   * @dbg_continuous: debug state variable to restore PRU execution mode
> >>   * @evt_count: number of mapped events
> >> + * @gpmux_save: saved value for gpmux config
> >>   */
> >>  struct pru_rproc {
> >>  	int id;
> >> @@ -141,6 +142,7 @@ struct pru_rproc {
> >>  	u32 dbg_single_step;
> >>  	u32 dbg_continuous;
> >>  	u8 evt_count;
> >> +	u8 gpmux_save;
> >>  };
> >>  
> >>  static inline u32 pru_control_read_reg(struct pru_rproc *pru, unsigned int reg)
> >> @@ -250,6 +252,7 @@ struct rproc *pru_rproc_get(struct device_node *np, int index,
> >>  	struct device *dev;
> >>  	const char *fw_name;
> >>  	int ret;
> >> +	u32 mux;
> >>  
> >>  	try_module_get(THIS_MODULE);
> >>  
> >> @@ -273,6 +276,22 @@ struct rproc *pru_rproc_get(struct device_node *np, int index,
> >>  
> >>  	mutex_unlock(&pru->lock);
> >>  
> >> +	ret = pruss_cfg_get_gpmux(pru->pruss, pru->id, &pru->gpmux_save);
> >> +	if (ret) {
> >> +		dev_err(dev, "failed to get cfg gpmux: %d\n", ret);
> >> +		goto err;
> >> +	}
> >> +
> >> +	ret = of_property_read_u32_index(np, "ti,pruss-gp-mux-sel", index,
> >> +					 &mux);
> >> +	if (!ret) {
> >> +		ret = pruss_cfg_set_gpmux(pru->pruss, pru->id, mux);
> >> +		if (ret) {
> >> +			dev_err(dev, "failed to set cfg gpmux: %d\n", ret);
> >> +			goto err;
> >> +		}
> >> +	}
> >> +
> >>  	if (pru_id)
> >>  		*pru_id = pru->id;
> >>  
> >> @@ -310,6 +329,7 @@ void pru_rproc_put(struct rproc *rproc)
> >>  
> >>  	pru = rproc->priv;
> >>  
> >> +	pruss_cfg_set_gpmux(pru->pruss, pru->id, pru->gpmux_save);
> >>  	pru_rproc_set_firmware(rproc, NULL);
> >>
> > 
> >   CC      drivers/remoteproc/pru_rproc.o
> > /home/mpoirier/work/remoteproc/kernel-review/drivers/remoteproc/pru_rproc.c: In function ‘pru_rproc_get’:
> > /home/mpoirier/work/remoteproc/kernel-review/drivers/remoteproc/pru_rproc.c:279:8: error: implicit declaration of function ‘pruss_cfg_get_gpmux’ [-Werror=implicit-function-declaration]
> >   279 |  ret = pruss_cfg_get_gpmux(pru->pruss, pru->id, &pru->gpmux_save);
> >       |        ^~~~~~~~~~~~~~~~~~~
> > /home/mpoirier/work/remoteproc/kernel-review/drivers/remoteproc/pru_rproc.c:288:9: error: implicit declaration of function ‘pruss_cfg_set_gpmux’ [-Werror=implicit-function-declaration]
> >   288 |   ret = pruss_cfg_set_gpmux(pru->pruss, pru->id, mux);
> >       |         ^~~~~~~~~~~~~~~~~~~
> > 
> > I get this on both rproc-next and today's linux next.  
> 
> This patch is dependent on the series [2] Introduce PRU platform consumer API
> https://patchwork.kernel.org/project/linux-remoteproc/cover/20220406094358.7895-1-p-mohan@ti.com/
> as the api pruss_cfg_get_gpmux calls the api pruss_cfg_read and the api
> pruss_cfg_set_gpmux calls the api pruss_cfg_update which are implemented by the
> patch "soc: ti: pruss: Add pruss_cfg_read()/update() API" in the above series.
>

What I read from the cover letter is that the series you are referring to was an
RFC that predated this series.  Also from the cover letter, the second paragraph
clearly indicate that two other series _depend_ on this series.  As such there
was no way for me to identify the dependency.  

> This error is coming as the dependent patch needs the patch "soc: ti: pruss:
> Add pruss_cfg_read()/update() API" to be applied for compilation.

Please provide a link when referencing patchsets.  That way we know exactly
which one we are talking about.

> 
> Thanks,
> Danish.
> 
> > 
> >>  	mutex_lock(&pru->lock);
> >> -- 
> >> 2.17.1
> >>

Re: [EXTERNAL] Re: [PATCH v5 6/6] remoteproc: pru: add support for configuring GPMUX based on client setup

[Md Danish Anwar]

Hi Mathieu

On 07/09/22 00:59, Mathieu Poirier wrote:
> Good day,
> 
> On Mon, Sep 05, 2022 at 02:41:27PM +0530, Md Danish Anwar wrote:
>> Hi Mathieu,
>>
>> On 19/07/22 21:46, Mathieu Poirier wrote:
>>> On Tue, Jun 07, 2022 at 10:26:50AM +0530, Puranjay Mohan wrote:
>>>> From: Tero Kristo <t-kristo@ti.com>
>>>>
>>>> Client device node property ti,pruss-gp-mux-sel can now be used to
>>>> configure the GPMUX config value for PRU.
>>>>
>>>> Signed-off-by: Tero Kristo <t-kristo@ti.com>
>>>> [s-anna@ti.com: simplify the pru id usage]
>>>> Signed-off-by: Suman Anna <s-anna@ti.com>
>>>> Signed-off-by: Puranjay Mohan <p-mohan@ti.com>
>>>> ---
>>>> V4->v5
>>>> * This patch was included in v4 and had some checkpatch errors that have
>>>>   been resolved in v5
>>>> ---
>>>>  drivers/remoteproc/pru_rproc.c | 20 ++++++++++++++++++++
>>>>  1 file changed, 20 insertions(+)
>>>>
>>>> diff --git a/drivers/remoteproc/pru_rproc.c b/drivers/remoteproc/pru_rproc.c
>>>> index 2977eb50631b..f2c6c55f0f20 100644
>>>> --- a/drivers/remoteproc/pru_rproc.c
>>>> +++ b/drivers/remoteproc/pru_rproc.c
>>>> @@ -123,6 +123,7 @@ struct pru_private_data {
>>>>   * @dbg_single_step: debug state variable to set PRU into single step mode
>>>>   * @dbg_continuous: debug state variable to restore PRU execution mode
>>>>   * @evt_count: number of mapped events
>>>> + * @gpmux_save: saved value for gpmux config
>>>>   */
>>>>  struct pru_rproc {
>>>>  	int id;
>>>> @@ -141,6 +142,7 @@ struct pru_rproc {
>>>>  	u32 dbg_single_step;
>>>>  	u32 dbg_continuous;
>>>>  	u8 evt_count;
>>>> +	u8 gpmux_save;
>>>>  };
>>>>  
>>>>  static inline u32 pru_control_read_reg(struct pru_rproc *pru, unsigned int reg)
>>>> @@ -250,6 +252,7 @@ struct rproc *pru_rproc_get(struct device_node *np, int index,
>>>>  	struct device *dev;
>>>>  	const char *fw_name;
>>>>  	int ret;
>>>> +	u32 mux;
>>>>  
>>>>  	try_module_get(THIS_MODULE);
>>>>  
>>>> @@ -273,6 +276,22 @@ struct rproc *pru_rproc_get(struct device_node *np, int index,
>>>>  
>>>>  	mutex_unlock(&pru->lock);
>>>>  
>>>> +	ret = pruss_cfg_get_gpmux(pru->pruss, pru->id, &pru->gpmux_save);
>>>> +	if (ret) {
>>>> +		dev_err(dev, "failed to get cfg gpmux: %d\n", ret);
>>>> +		goto err;
>>>> +	}
>>>> +
>>>> +	ret = of_property_read_u32_index(np, "ti,pruss-gp-mux-sel", index,
>>>> +					 &mux);
>>>> +	if (!ret) {
>>>> +		ret = pruss_cfg_set_gpmux(pru->pruss, pru->id, mux);
>>>> +		if (ret) {
>>>> +			dev_err(dev, "failed to set cfg gpmux: %d\n", ret);
>>>> +			goto err;
>>>> +		}
>>>> +	}
>>>> +
>>>>  	if (pru_id)
>>>>  		*pru_id = pru->id;
>>>>  
>>>> @@ -310,6 +329,7 @@ void pru_rproc_put(struct rproc *rproc)
>>>>  
>>>>  	pru = rproc->priv;
>>>>  
>>>> +	pruss_cfg_set_gpmux(pru->pruss, pru->id, pru->gpmux_save);
>>>>  	pru_rproc_set_firmware(rproc, NULL);
>>>>
>>>
>>>   CC      drivers/remoteproc/pru_rproc.o
>>> /home/mpoirier/work/remoteproc/kernel-review/drivers/remoteproc/pru_rproc.c: In function ‘pru_rproc_get’:
>>> /home/mpoirier/work/remoteproc/kernel-review/drivers/remoteproc/pru_rproc.c:279:8: error: implicit declaration of function ‘pruss_cfg_get_gpmux’ [-Werror=implicit-function-declaration]
>>>   279 |  ret = pruss_cfg_get_gpmux(pru->pruss, pru->id, &pru->gpmux_save);
>>>       |        ^~~~~~~~~~~~~~~~~~~
>>> /home/mpoirier/work/remoteproc/kernel-review/drivers/remoteproc/pru_rproc.c:288:9: error: implicit declaration of function ‘pruss_cfg_set_gpmux’ [-Werror=implicit-function-declaration]
>>>   288 |   ret = pruss_cfg_set_gpmux(pru->pruss, pru->id, mux);
>>>       |         ^~~~~~~~~~~~~~~~~~~
>>>
>>> I get this on both rproc-next and today's linux next.  
>>
>> This patch is dependent on the series [2] Introduce PRU platform consumer API
>> https://patchwork.kernel.org/project/linux-remoteproc/cover/20220406094358.7895-1-p-mohan@ti.com/
>> as the api pruss_cfg_get_gpmux calls the api pruss_cfg_read and the api
>> pruss_cfg_set_gpmux calls the api pruss_cfg_update which are implemented by the
>> patch "soc: ti: pruss: Add pruss_cfg_read()/update() API" in the above series.
>>
> 
> What I read from the cover letter is that the series you are referring to was an
> RFC that predated this series.  Also from the cover letter, the second paragraph
> clearly indicate that two other series _depend_ on this series.  As such there
> was no way for me to identify the dependency.  
> 
>> This error is coming as the dependent patch needs the patch "soc: ti: pruss:
>> Add pruss_cfg_read()/update() API" to be applied for compilation.
> 
> Please provide a link when referencing patchsets.  That way we know exactly
> which one we are talking about.
> 

Sure, the current patch is dependent on the patch [PATCH v2 3/6] soc: ti:
pruss: Add pruss_cfg_read()/update() API
https://lore.kernel.org/lkml/20220418123004.9332-4-p-mohan@ti.com/

Thanks,
Danish.

>>
>> Thanks,
>> Danish.
>>
>>>
>>>>  	mutex_lock(&pru->lock);
>>>> -- 
>>>> 2.17.1
>>>>