Skip to contentSkip to conversation

ACPI _CRS CSI-2 and MIPI DisCo for Imaging support

PATCH Superseded Sakari Ailus 3 participants 23 messages Jan 25, 2023 – Feb 2, 2023

Subsystems (5)

v3 of 8 Superseded Latest is v8 →

lore
Share & tools
lore

Sakari Ailus

author reviewer
#0 lore raw

v3 cover ACPI _CRS CSI-2 and MIPI DisCo for Imaging support

Hello all,

Here's an implementation of ACPI 6.4 _CRS CSI-2 resource descriptor and MIPI DisCo for Imaging 1.0 [1]. What the two basically provide is an officially sanctioned way to describe CSI-2 connected cameras to operating system software, something DT based systems have enjoyed for quite some time already.

The implementation digs the information from ACPI tables (_CRS descriptors and data + property extensions) and constructs software nodes that are compatible with Documentation/firmware-guide/acpi/dsd/graph.rst and Documentation/devicetree/bindings/media/video-interface-devices.yaml . No specific driver changes are needed.

These patches are on the testing branch of the linux-acpi tree where they depend on the patch constifying the ACPI pathname argument for acpi_get_handle() (commit 91fdb91ccca2b48572a1ccf1d382fd599e3e1237).

[1] https://www.mipi.org/specifications/mipi-disco-imaging

since v2:

- Unwrap a few lines.

- Copy CSI-2 resource source string using strscpy() instead of memcpy() in scan_check_crs_csi2_instance.

- Fix GRAPH_PORT_NAME() sanity check bug introduced in v2.

- Fix snprintf() return value check for port node name in get_mipi_port_handle().

- Fix mipi-img-lane-polarities reading.

- Cast bit value to bool instead of using ... ? 1U : 0U.

- Get primary fwnode using acpi_fwnode_handle().

- Don't use MIPI_IMG_PREFIX in the array of renamed properties.

- Use tabs for indenting drivers/acpi/property.c authors.

- Add a comment on assigning ACPI device's secondary fwnode and assign ACPI device's secondary fwnode straight to NULL when unassigning it.

since v1:

- Update copyright notices.

- Include linux/types.h instead of linux/kernel.h in drivers/acpi/mipi.c.

- Use SWNODE_GRAPH_PORT_NAME_FMT instead of plain "port@%u" in GRAPH_PORT_NAME macro.

- Make the condition in NEXT_PROPERTY() macro easier to read.

- Unwrap lines to make them moderately longer than 80 characters.

- Use * BITS_PER_TYPE(u8) instead of << 3 to convert bytes to bits in init_port_csi2_common().

- Test ACPI framework call success using ACPI_SUCCESS() instead of comparing with AE_OK. Likewise for ACPI_FAILURE and != AE_OK.

- Use newly added SOFTWARE_NODE() macro to construct the root software node.

- Use str_has_prefix() to test for a string prefix instead of memcmp().

- Add pr_fmt() macro to drivers/acpi/property.c.

- Move logical or operators to the end of the line in acpi_properties_prepare().

- Improve bad node type error in acpi_parse_string_ref().

Sakari Ailus (8):
ACPI: property: Parse data node string references in properties
ACPI: property: Parse _CRS CSI-2 descriptor
device property: Add SOFTWARE_NODE() macro for defining software nodes
ACPI: property: Generate camera swnodes for ACPI and DisCo for Imaging
ACPI: property: Dig "rotation" property for devices with CSI2 _CRS
ACPI: property: Rename parsed MIPI DisCo for Imaging properties
ACPI: property: Skip MIPI property table without "mipi-img" prefix
ACPI: property: Document _CRS CSI-2 and DisCo for Imaging support

 drivers/acpi/Makefile    |   2 +-
 drivers/acpi/internal.h  |   9 +
 drivers/acpi/mipi.c      | 768 +++++++++++++++++++++++++++++++++++++++
 drivers/acpi/property.c  | 132 +++++--
 drivers/acpi/scan.c      |  35 +-
 include/acpi/acpi_bus.h  |  61 ++++
 include/linux/property.h |   7 +
 7 files changed, 983 insertions(+), 31 deletions(-)
 create mode 100644 drivers/acpi/mipi.c
Signature · 1 line
2.30.2
author reviewer
#1 lore raw

PATCH v3 1/8 ACPI: property: Parse data node string references in properties

Add support for parsing property references using strings, besides reference objects that were previously supported. This allows also referencing data nodes which was not possible with reference objects.

Also add pr_fmt() macro to prefix printouts.

While at it, update copyright.

---
 drivers/acpi/property.c | 110 ++++++++++++++++++++++++++++++++++------
 1 file changed, 94 insertions(+), 16 deletions(-)
drivers/acpi/property.c +94 -16
index b8d9eb9a433ed..2b69b25c405c5 100644
@@ -2,14 +2,17 @@
 /*
  * ACPI device specific properties support.
  *
- * Copyright (C) 2014, Intel Corporation
+ * Copyright (C) 2014--2023, Intel Corporation
  * All rights reserved.
  *
  * Authors: Mika Westerberg <[email protected]>
- *          Darren Hart <[email protected]>
- *          Rafael J. Wysocki <[email protected]>
+ *	    Darren Hart <[email protected]>
+ *	    Rafael J. Wysocki <[email protected]>
+ *	    Sakari Ailus <[email protected]>
  */
 
+#define pr_fmt(fmt) "ACPI: " fmt
+
 #include <linux/acpi.h>
 #include <linux/device.h>
 #include <linux/export.h>
@@ -795,7 +798,8 @@ acpi_fwnode_get_named_child_node(const struct fwnode_handle *fwnode,
 static int acpi_get_ref_args(struct fwnode_reference_args *args,
 			     struct fwnode_handle *ref_fwnode,
 			     const union acpi_object **element,
-			     const union acpi_object *end, size_t num_args)
+			     const union acpi_object *end, size_t num_args,
+			     bool subnode_string)
 {
 	u32 nargs = 0, i;
 
@@ -803,13 +807,16 @@ static int acpi_get_ref_args(struct fwnode_reference_args *args,
 	 * Find the referred data extension node under the
 	 * referred device node.
 	 */
-	for (; *element < end && (*element)->type == ACPI_TYPE_STRING;
-	     (*element)++) {
-		const char *child_name = (*element)->string.pointer;
-
-		ref_fwnode = acpi_fwnode_get_named_child_node(ref_fwnode, child_name);
-		if (!ref_fwnode)
-			return -EINVAL;
+	if (subnode_string) {
+		for (; *element < end && (*element)->type == ACPI_TYPE_STRING;
+		     (*element)++) {
+			const char *child_name = (*element)->string.pointer;
+
+			ref_fwnode = acpi_fwnode_get_named_child_node(ref_fwnode,
+								      child_name);
+			if (!ref_fwnode)
+				return -EINVAL;
+		}
 	}
 
 	/*
@@ -820,7 +827,8 @@ static int acpi_get_ref_args(struct fwnode_reference_args *args,
 	for (i = 0; (*element) + i < end && i < num_args; i++) {
 		acpi_object_type type = (*element)[i].type;
 
-		if (type == ACPI_TYPE_LOCAL_REFERENCE)
+		if (type == ACPI_TYPE_LOCAL_REFERENCE ||
+		    (!subnode_string && type == ACPI_TYPE_STRING))
 			break;
 
 		if (type == ACPI_TYPE_INTEGER)
@@ -844,6 +852,43 @@ static int acpi_get_ref_args(struct fwnode_reference_args *args,
 	return 0;
 }
 
+static struct fwnode_handle *
+acpi_parse_string_ref(const struct fwnode_handle *fwnode, const char *refstring)
+{
+	acpi_handle scope, handle;
+	struct acpi_data_node *dn;
+	struct acpi_device *device;
+	acpi_status status;
+
+	if (is_acpi_device_node(fwnode)) {
+		scope = to_acpi_device_node(fwnode)->handle;
+	} else if (is_acpi_data_node(fwnode)) {
+		scope = to_acpi_data_node(fwnode)->handle;
+	} else {
+		pr_err("bad node type for node %pfw\n", fwnode);
+		return ERR_PTR(-EINVAL);
+	}
+
+	status = acpi_get_handle(scope, refstring, &handle);
+	if (ACPI_FAILURE(status)) {
+		acpi_handle_debug(scope, "can't get handle for %s", refstring);
+		return ERR_PTR(-EINVAL);
+	}
+
+	device = acpi_fetch_acpi_dev(handle);
+	if (device)
+		return acpi_fwnode_handle(device);
+
+	status = acpi_get_data_full(handle, acpi_nondev_subnode_tag,
+				    (void **)&dn, NULL);
+	if (ACPI_FAILURE(status) || !dn) {
+		acpi_handle_debug(handle, "can't find subnode");
+		return ERR_PTR(-EINVAL);
+	}
+
+	return &dn->fwnode;
+}
+
 /**
  * __acpi_node_get_property_reference - returns handle to the referenced object
  * @fwnode: Firmware node to get the property from
@@ -886,6 +931,7 @@ int __acpi_node_get_property_reference(const struct fwnode_handle *fwnode,
 	const union acpi_object *element, *end;
 	const union acpi_object *obj;
 	const struct acpi_device_data *data;
+	struct fwnode_handle *ref_fwnode;
 	struct acpi_device *device;
 	int ret, idx = 0;
 
@@ -909,16 +955,29 @@ int __acpi_node_get_property_reference(const struct fwnode_handle *fwnode,
 
 		args->fwnode = acpi_fwnode_handle(device);
 		args->nargs = 0;
+		return 0;
+	case ACPI_TYPE_STRING:
+		if (index)
+			return -ENOENT;
+
+		ref_fwnode = acpi_parse_string_ref(fwnode, obj->string.pointer);
+		if (IS_ERR(ref_fwnode))
+			return PTR_ERR(ref_fwnode);
+
+		args->fwnode = ref_fwnode;
+		args->nargs = 0;
+
 		return 0;
 	case ACPI_TYPE_PACKAGE:
 		/*
 		 * If it is not a single reference, then it is a package of
-		 * references followed by number of ints as follows:
+		 * references, followed by number of ints as follows:
 		 *
 		 *  Package () { REF, INT, REF, INT, INT }
 		 *
-		 * The index argument is then used to determine which reference
-		 * the caller wants (along with the arguments).
+		 * Here, REF may be either a local reference or a string. The
+		 * index argument is then used to determine which reference the
+		 * caller wants (along with the arguments).
 		 */
 		break;
 	default:
@@ -942,7 +1001,26 @@ int __acpi_node_get_property_reference(const struct fwnode_handle *fwnode,
 
 			ret = acpi_get_ref_args(idx == index ? args : NULL,
 						acpi_fwnode_handle(device),
-						&element, end, num_args);
+						&element, end, num_args, true);
+			if (ret < 0)
+				return ret;
+
+			if (idx == index)
+				return 0;
+
+			break;
+		case ACPI_TYPE_STRING:
+			ref_fwnode =
+				acpi_parse_string_ref(fwnode,
+						      element->string.pointer);
+			if (IS_ERR(ref_fwnode))
+				return PTR_ERR(ref_fwnode);
+
+			element++;
+
+			ret = acpi_get_ref_args(idx == index ? args : NULL,
+						ref_fwnode, &element, end,
+						num_args, false);
 			if (ret < 0)
 				return ret;
 
Signature · 1 line
2.30.2
#2 lore raw

Re: [PATCH v3 1/8] ACPI: property: Parse data node string references in properties

Quoted 7 lines · Sakari Ailus

Add support for parsing property references using strings, besides reference objects that were previously supported. This allows also referencing data nodes which was not possible with reference objects.

Also add pr_fmt() macro to prefix printouts.

While at it, update copyright.

...

Quoted 2 lines

- * Copyright (C) 2014, Intel Corporation
+ * Copyright (C) 2014--2023, Intel Corporation

As I pointed out this is non-traditional way on how you provide range of the years in the Copyright line.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#3 lore raw

Re: [PATCH v3 1/8] ACPI: property: Parse data node string references in properties

In reply to Sakari Ailus
Quoted 10 lines · Sakari Ailus

Add support for parsing property references using strings, besides reference objects that were previously supported. This allows also referencing data nodes which was not possible with reference objects.

Also add pr_fmt() macro to prefix printouts.

While at it, update copyright.

While there is some design documentation in the last patch, at least some pieces of it should go into the changelog of this patch IMO.

Also the spec-related information from the cover letter would be useful here, because cover letters don't go into the git record.

#4 lore raw

Re: [PATCH v3 1/8] ACPI: property: Parse data node string references in properties

Quoted 9 lines · Rafael J. Wysocki
Quoted 7 lines · Sakari Ailus

Add support for parsing property references using strings, besides reference objects that were previously supported. This allows also referencing data nodes which was not possible with reference objects.

Also add pr_fmt() macro to prefix printouts.

Quoted 7 lines
Quoted 1 line

While at it, update copyright.

While there is some design documentation in the last patch, at least some pieces of it should go into the changelog of this patch IMO.

Also the spec-related information from the cover letter would be useful here, because cover letters don't go into the git record.

Interestingly if a maintainer starts using b4 tool it has the mode in which the series is represented as a pull request and cover letter contents go as a commit message into the merge commit. That said, it depends on the maintainer's workflow.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
author reviewer
#5 lore raw

PATCH v3 2/8 ACPI: property: Parse _CRS CSI-2 descriptor

In reply to Sakari Ailus

Parse newly added ACPI _CRS CSI-2 descriptor for CSI-2 and camera configuration. For now, only figure out where the descriptor is present in order to allow adding information from it to related devices.

---
 drivers/acpi/Makefile   |   2 +-
 drivers/acpi/internal.h |   7 +
 drivers/acpi/mipi.c     | 320 ++++++++++++++++++++++++++++++++++++++++
 drivers/acpi/scan.c     |  16 +-
 include/acpi/acpi_bus.h |  11 ++
 5 files changed, 351 insertions(+), 5 deletions(-)
 create mode 100644 drivers/acpi/mipi.c
drivers/acpi/Makefile +1 -1
index a5b649e71ab1b..a98fa1bc15548 100644
@@ -37,7 +37,7 @@ acpi-$(CONFIG_ACPI_SLEEP)	+= proc.o
 # ACPI Bus and Device Drivers
 #
 acpi-y				+= bus.o glue.o
-acpi-y				+= scan.o
+acpi-y				+= scan.o mipi.o
 acpi-y				+= resource.o
 acpi-y				+= acpi_processor.o
 acpi-y				+= processor_core.o
drivers/acpi/internal.h +7 -0
index ec584442fb298..1ec4aa92bf17a 100644
@@ -282,4 +282,11 @@ void acpi_init_lpit(void);
 static inline void acpi_init_lpit(void) { }
 #endif
 
+/*--------------------------------------------------------------------------
+				ACPI and MIPI DisCo for Imaging conversion
+  -------------------------------------------------------------------------- */
+
+void acpi_crs_csi2_swnodes_del_free(void);
+void acpi_bus_scan_crs_csi2(acpi_handle handle);
+
 #endif /* _ACPI_INTERNAL_H_ */
drivers/acpi/mipi.c new file +320 -0

320 lines not on this page · raw message

drivers/acpi/scan.c +12 -4
index 0c6f06abe3f47..50de874b8f208 100644
@@ -2423,9 +2423,12 @@ EXPORT_SYMBOL_GPL(acpi_dev_get_next_consumer_dev);
 int acpi_bus_scan(acpi_handle handle)
 {
 	struct acpi_device *device = NULL;
+	int ret = 0;
 
 	acpi_bus_scan_second_pass = false;
 
+	acpi_bus_scan_crs_csi2(handle);
+
 	/* Pass 1: Avoid enumerating devices with missing dependencies. */
 
 	if (ACPI_SUCCESS(acpi_bus_check_add(handle, true, &device)))
@@ -2433,13 +2436,15 @@ int acpi_bus_scan(acpi_handle handle)
 				    acpi_bus_check_add_1, NULL, NULL,
 				    (void **)&device);
 
-	if (!device)
-		return -ENODEV;
+	if (!device) {
+		ret = -ENODEV;
+		goto out_release;
+	}
 
 	acpi_bus_attach(device, (void *)true);
 
 	if (!acpi_bus_scan_second_pass)
-		return 0;
+		goto out_release;
 
 	/* Pass 2: Enumerate all of the remaining devices. */
 
@@ -2452,7 +2457,10 @@ int acpi_bus_scan(acpi_handle handle)
 
 	acpi_bus_attach(device, NULL);
 
-	return 0;
+out_release:
+	acpi_crs_csi2_swnodes_del_free();
+
+	return ret;
 }
 EXPORT_SYMBOL(acpi_bus_scan);
 
include/acpi/acpi_bus.h +11 -0
index e44be31115a67..a05fe22c1175c 100644
@@ -360,6 +360,17 @@ struct acpi_device_data {
 
 struct acpi_gpio_mapping;
 
+struct acpi_device_software_node_port {
+	unsigned int port_nr;
+};
+
+struct acpi_device_software_nodes {
+	struct acpi_device_software_node_port *ports;
+	struct software_node *nodes;
+	const struct software_node **nodeptrs;
+	unsigned int num_ports;
+};
+
 /* Device */
 struct acpi_device {
 	u32 pld_crc;
Signature · 1 line
2.30.2
#6 lore raw

Re: [PATCH v3 2/8] ACPI: property: Parse _CRS CSI-2 descriptor

Quoted 3 lines · Sakari Ailus

Parse newly added ACPI _CRS CSI-2 descriptor for CSI-2 and camera configuration. For now, only figure out where the descriptor is present in order to allow adding information from it to related devices.

...

drivers/acpi/mipi.c line 84

Quoted 1 line
drivers/acpi/mipi.c new file +1 -0
+	if (status != AE_OK) {

ACPI_FAILURE()

Quoted 5 lines
drivers/acpi/mipi.c new file +5 -0
+		acpi_handle_warn(inst_context->handle,
+				 "cannot get handle for %s\n",
+				 csi2->resource_source.string_ptr);
+		return AE_OK;
+	}

...

drivers/acpi/mipi.c line 118

Quoted 8 lines
drivers/acpi/mipi.c new file +7 -0
+	struct scan_check_crs_csi2_context inst_context = {
+		.handle = handle,
+		.res_list = LIST_HEAD_INIT(inst_context.res_list),
+	};
+	struct list_head *list = context;
+	struct crs_csi2 *csi2;
 
+	INIT_LIST_HEAD(&inst_context.res_list);

Do you still need this?

...

drivers/acpi/mipi.c line 124

Quoted 5 lines
drivers/acpi/mipi.c new file +5 -0
+	acpi_walk_resources(handle, METHOD_NAME__CRS,
+			    scan_check_crs_csi2_instance, &inst_context);
+
+	if (list_empty(&inst_context.res_list))
+		return AE_OK;

I'm wondering if you can utilize acpi_dev_get_resources().

...

drivers/acpi/mipi.c line 201

Quoted 2 lines
drivers/acpi/mipi.c new file +2 -0
+	/* Collect the devices that have a _CRS CSI-2 resource */
+	acpi_walk_namespace(ACPI_TYPE_DEVICE, handle, ACPI_UINT32_MAX,

Other serial buses limit the DEPTH by 32, why do we need more here?

Quoted 1 line
drivers/acpi/mipi.c new file +1 -0
+			    scan_check_crs_csi2, NULL, &crs_csi2_handles, NULL);

...

drivers/acpi/mipi.c line 248

Quoted 2 lines
drivers/acpi/mipi.c new file +2 -0
+	sort(handle_refs, handle_count, sizeof(*handle_refs), crs_handle_cmp,
+	     NULL);

A single line?

...

drivers/acpi/mipi.c line 276

Quoted 4 lines
drivers/acpi/mipi.c new file +4 -0
+		if (check_mul_overflow(sizeof(*ads->ports) +
+				       sizeof(*ads->nodes) * 2 +
+				       sizeof(*ads->nodeptrs) * 2,
+				       (size_t)this_count, &alloc_size) ||

So, now you know why this_count can't be type of size_t?

Quoted 7 lines
drivers/acpi/mipi.c new file +7 -0
+		    check_add_overflow(sizeof(*ads) + sizeof(*ads->nodes) +
+				       sizeof(*ads->nodeptrs) * 2,
+				       alloc_size, &alloc_size)) {
+			acpi_handle_warn(handle, "too many handles (%u)",
+					 this_count);
+			continue;
+		}
Signature · 2 lines
With Best Regards,
Andy Shevchenko
author reviewer
#7 lore raw

Re: [PATCH v3 2/8] ACPI: property: Parse _CRS CSI-2 descriptor

Hi Andy,

Thank you for the review.

Quoted 10 lines · Andy Shevchenko
Quoted 3 lines · Sakari Ailus

Parse newly added ACPI _CRS CSI-2 descriptor for CSI-2 and camera configuration. For now, only figure out where the descriptor is present in order to allow adding information from it to related devices.

...

drivers/acpi/mipi.c line 84

Quoted 1 line
drivers/acpi/mipi.c new file +1 -0
+	if (status != AE_OK) {

ACPI_FAILURE()

Yes.

Quoted 16 lines
Quoted 6 lines
drivers/acpi/mipi.c new file +5 -0
+		acpi_handle_warn(inst_context->handle,
+				 "cannot get handle for %s\n",
+				 csi2->resource_source.string_ptr);
+		return AE_OK;
+	}

...

Quoted 6 lines
drivers/acpi/mipi.c new file +6 -0
+	struct scan_check_crs_csi2_context inst_context = {
+		.handle = handle,
+		.res_list = LIST_HEAD_INIT(inst_context.res_list),
+	};
+	struct list_head *list = context;
+	struct crs_csi2 *csi2;
Quoted 3 lines

drivers/acpi/mipi.c line 118

Quoted 1 line
drivers/acpi/mipi.c new file +1 -0
+	INIT_LIST_HEAD(&inst_context.res_list);

Do you still need this?

Oops. Forgot to remove it, I'll do in v4.

Quoted 10 lines

...

drivers/acpi/mipi.c line 124

Quoted 5 lines
drivers/acpi/mipi.c new file +5 -0
+	acpi_walk_resources(handle, METHOD_NAME__CRS,
+			    scan_check_crs_csi2_instance, &inst_context);
+
+	if (list_empty(&inst_context.res_list))
+		return AE_OK;

I'm wondering if you can utilize acpi_dev_get_resources().

We don't have an acpi_device yet. I'd rather keep scanning for _CRS CSI2 resources here, as we'd otherwise have to split creating acpi_device's and registering them into two.

I'd say if someone had implemented this like that, I'd ask them to change it.

Quoted 7 lines

...

drivers/acpi/mipi.c line 201

Quoted 2 lines
drivers/acpi/mipi.c new file +2 -0
+	/* Collect the devices that have a _CRS CSI-2 resource */
+	acpi_walk_namespace(ACPI_TYPE_DEVICE, handle, ACPI_UINT32_MAX,

Other serial buses limit the DEPTH by 32, why do we need more here?

I'm using the same value as for scanning devices to be added. Effectively this is not a practical limit but it doesn't need to be.

Quoted 9 lines
Quoted 2 lines
drivers/acpi/mipi.c new file +1 -0
+			    scan_check_crs_csi2, NULL, &crs_csi2_handles, NULL);

...

drivers/acpi/mipi.c line 248

Quoted 2 lines
drivers/acpi/mipi.c new file +2 -0
+	sort(handle_refs, handle_count, sizeof(*handle_refs), crs_handle_cmp,
+	     NULL);

A single line?

Can do...

Quoted 9 lines

...

drivers/acpi/mipi.c line 276

Quoted 4 lines
drivers/acpi/mipi.c new file +4 -0
+		if (check_mul_overflow(sizeof(*ads->ports) +
+				       sizeof(*ads->nodes) * 2 +
+				       sizeof(*ads->nodeptrs) * 2,
+				       (size_t)this_count, &alloc_size) ||

So, now you know why this_count can't be type of size_t?

Forgot to change this one, thanks!

Quoted 9 lines
Quoted 8 lines
drivers/acpi/mipi.c new file +7 -0
+		    check_add_overflow(sizeof(*ads) + sizeof(*ads->nodes) +
+				       sizeof(*ads->nodeptrs) * 2,
+				       alloc_size, &alloc_size)) {
+			acpi_handle_warn(handle, "too many handles (%u)",
+					 this_count);
+			continue;
+		}
Signature · 3 lines
Kind regards,

Sakari Ailus
#8 lore raw

Re: [PATCH v3 2/8] ACPI: property: Parse _CRS CSI-2 descriptor

Quoted 2 lines · Sakari Ailus
Quoted 1 line · Andy Shevchenko

On Thu, Jan 26, 2023 at 12:40:55AM +0200, Sakari Ailus wrote:

...

Quoted 7 lines
Quoted 4 lines

drivers/acpi/mipi.c line 201

Quoted 2 lines
drivers/acpi/mipi.c new file +2 -0
+	/* Collect the devices that have a _CRS CSI-2 resource */
+	acpi_walk_namespace(ACPI_TYPE_DEVICE, handle, ACPI_UINT32_MAX,

Other serial buses limit the DEPTH by 32, why do we need more here?

I'm using the same value as for scanning devices to be added. Effectively this is not a practical limit but it doesn't need to be.

Hmm... Theoretically one may supply a malformed table, quite compact, but with huge amount of "fake" devices. I would limit this the same way. If we got a report of the depth not being enough we may bump based on the reality. This also may limit scanning time.

Quoted 1 line
Quoted 1 line
Quoted 1 line
drivers/acpi/mipi.c new file +1 -0
+			    scan_check_crs_csi2, NULL, &crs_csi2_handles, NULL);
Signature · 2 lines
With Best Regards,
Andy Shevchenko
author reviewer
#9 lore raw

Re: [PATCH v3 2/8] ACPI: property: Parse _CRS CSI-2 descriptor

Quoted 11 lines · Andy Shevchenko
Quoted 2 lines · Sakari Ailus
Quoted 1 line · Andy Shevchenko

On Thu, Jan 26, 2023 at 12:40:55AM +0200, Sakari Ailus wrote:

...

Quoted 5 lines
Quoted 4 lines

drivers/acpi/mipi.c line 201

Quoted 2 lines
drivers/acpi/mipi.c new file +2 -0
+	/* Collect the devices that have a _CRS CSI-2 resource */
+	acpi_walk_namespace(ACPI_TYPE_DEVICE, handle, ACPI_UINT32_MAX,

Other serial buses limit the DEPTH by 32, why do we need more here?

Quoted 7 lines
Quoted 2 lines

I'm using the same value as for scanning devices to be added. Effectively this is not a practical limit but it doesn't need to be.

Hmm... Theoretically one may supply a malformed table, quite compact, but with huge amount of "fake" devices. I would limit this the same way. If we got a report of the depth not being enough we may bump based on the reality. This also may limit scanning time.

Then this should be done also for scanning ACPI devices below.

I'd like to have Rafael's opinion before making such changes. I'll prepend such a patch to the series in that case.

Signature · 1 line
Sakari Ailus
#10 lore raw

Re: [PATCH v3 2/8] ACPI: property: Parse _CRS CSI-2 descriptor

In reply to Sakari Ailus
Quoted 97 lines · Sakari Ailus

Parse newly added ACPI _CRS CSI-2 descriptor for CSI-2 and camera configuration. For now, only figure out where the descriptor is present in order to allow adding information from it to related devices.

---
 drivers/acpi/Makefile   |   2 +-
 drivers/acpi/internal.h |   7 +
 drivers/acpi/mipi.c     | 320 ++++++++++++++++++++++++++++++++++++++++
 drivers/acpi/scan.c     |  16 +-
 include/acpi/acpi_bus.h |  11 ++
 5 files changed, 351 insertions(+), 5 deletions(-)
 create mode 100644 drivers/acpi/mipi.c
drivers/acpi/Makefile +1 -1
index a5b649e71ab1b..a98fa1bc15548 100644
@@ -37,7 +37,7 @@ acpi-$(CONFIG_ACPI_SLEEP)     += proc.o
 # ACPI Bus and Device Drivers
 #
 acpi-y                         += bus.o glue.o
-acpi-y                         += scan.o
+acpi-y                         += scan.o mipi.o
 acpi-y                         += resource.o
 acpi-y                         += acpi_processor.o
 acpi-y                         += processor_core.o
drivers/acpi/internal.h +7 -0
index ec584442fb298..1ec4aa92bf17a 100644
@@ -282,4 +282,11 @@ void acpi_init_lpit(void);
 static inline void acpi_init_lpit(void) { }
 #endif
 
+/*--------------------------------------------------------------------------
+                               ACPI and MIPI DisCo for Imaging conversion
+  -------------------------------------------------------------------------- */
+
+void acpi_crs_csi2_swnodes_del_free(void);
+void acpi_bus_scan_crs_csi2(acpi_handle handle);
+
 #endif /* _ACPI_INTERNAL_H_ */
drivers/acpi/mipi.c new file +47 -0
new file mode 100644
index 0000000000000..269a6965d56e6
@@ -0,0 +1,320 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * MIPI DisCo for Imaging support.
+ *
+ * Copyright (C) 2023 Intel Corporation
+ */
+
+#include <linux/acpi.h>
+#include <linux/list.h>
+#include <linux/module.h>
+#include <linux/overflow.h>
+#include <linux/types.h>
+#include <linux/slab.h>
+#include <linux/sort.h>
+#include <linux/string.h>
+
+#include "internal.h"
+
+struct crs_csi2_swnodes {
+       struct list_head list;
+       acpi_handle handle;
+       struct acpi_device_software_nodes *ads;
+};
+
+static LIST_HEAD(crs_csi2_swnodes);
+
+static void crs_csi2_swnode_del_free(struct crs_csi2_swnodes *swnodes)
+{
+       list_del(&swnodes->list);
+       kfree(swnodes);
+}
+
+void acpi_crs_csi2_swnodes_del_free(void)
+{
+       struct crs_csi2_swnodes *swnodes, *swnodes_tmp;
+
+       list_for_each_entry_safe(swnodes, swnodes_tmp, &crs_csi2_swnodes, list)
+               crs_csi2_swnode_del_free(swnodes);
+}
+
+struct crs_csi2_instance {
+       struct list_head list;
+       struct acpi_resource_csi2_serialbus csi2;
+       acpi_handle remote_handle;
+       char remote_name[];
+};
+

drivers/acpi/mipi.c line 57

Quoted 10 lines
drivers/acpi/mipi.c new file +10 -0
+struct crs_csi2 {
+       struct list_head list;
+       acpi_handle handle;
+       struct list_head buses;
+};
+
+struct scan_check_crs_csi2_context {
+       struct list_head res_list;
+       acpi_handle handle;
+};

All of the structures above need to be described, including what they represent and what they are for.

Quoted 22 lines
drivers/acpi/mipi.c new file +22 -0
+
+static acpi_status scan_check_crs_csi2_instance(struct acpi_resource *res,
+                                               void *context)
+{
+       struct scan_check_crs_csi2_context *inst_context = context;
+       struct acpi_resource_csi2_serialbus *csi2;
+       struct crs_csi2_instance *inst;
+       acpi_handle remote_handle;
+       acpi_status status;
+
+       if (res->type != ACPI_RESOURCE_TYPE_SERIAL_BUS)
+               return AE_OK;
+
+       csi2 = &res->data.csi2_serial_bus;
+
+       if (csi2->type != ACPI_RESOURCE_SERIAL_TYPE_CSI2)
+               return AE_OK;
+
+       if (!csi2->resource_source.string_length) {
+               acpi_handle_debug(inst_context->handle,
+                                 "invalid resource source string length\n");
+               return AE_OK;

drivers/acpi/mipi.c line 89

Quoted 10 lines
drivers/acpi/mipi.c new file +10 -0
+       }
+
+       status = acpi_get_handle(NULL, csi2->resource_source.string_ptr,
+                                &remote_handle);
+       if (status != AE_OK) {
+               acpi_handle_warn(inst_context->handle,
+                                "cannot get handle for %s\n",
+                                csi2->resource_source.string_ptr);
+               return AE_OK;
+       }

After this, remote_handle will be the ACPI handle of an ACPI object representing a remote CSI-2 device IIUC.

drivers/acpi/mipi.c line 91

Quoted 2 lines
drivers/acpi/mipi.c new file +2 -0
+
+       inst = kmalloc(struct_size(inst, remote_name,

This line break is confusing.

drivers/acpi/mipi.c line 97

Quoted 6 lines
drivers/acpi/mipi.c new file +6 -0
+                                  csi2->resource_source.string_length),
+                      GFP_KERNEL);
+       if (!inst)
+               return AE_OK;
+
+       inst->csi2 = *csi2;

A pointer to the entire csi2 structure is stored in ints, so why is it still necessary to copy remote_name and remote_handle extracted from it to the additional fields?

drivers/acpi/mipi.c line 103

Quoted 6 lines
drivers/acpi/mipi.c new file +6 -0
+       strscpy(inst->remote_name, csi2->resource_source.string_ptr,
+               csi2->resource_source.string_length);
+       inst->csi2.resource_source.string_ptr = inst->remote_name;
+       inst->remote_handle = remote_handle;
+
+       list_add(&inst->list, &inst_context->res_list);

So this adds an entry to a list of "receivers" pointed to by the _CRS for one "sender"?

Quoted 11 lines
drivers/acpi/mipi.c new file +11 -0
+
+       return AE_OK;
+}
+
+static acpi_status scan_check_crs_csi2(acpi_handle handle, u32 nesting_level,
+                                      void *context, void **ret)
+{
+       struct scan_check_crs_csi2_context inst_context = {
+               .handle = handle,
+               .res_list = LIST_HEAD_INIT(inst_context.res_list),
+       };

drivers/acpi/mipi.c line 124

Quoted 10 lines
drivers/acpi/mipi.c new file +10 -0
+       struct list_head *list = context;
+       struct crs_csi2 *csi2;
+
+       INIT_LIST_HEAD(&inst_context.res_list);
+
+       acpi_walk_resources(handle, METHOD_NAME__CRS,
+                           scan_check_crs_csi2_instance, &inst_context);
+
+       if (list_empty(&inst_context.res_list))
+               return AE_OK;

At this point inst_context.res_list is a list of remote "receivers" pointed to by all of the CSI-2 _CRS entries for the device represented by @handle?

drivers/acpi/mipi.c line 131

Quoted 7 lines
drivers/acpi/mipi.c new file +7 -0
+
+       csi2 = kmalloc(sizeof(*csi2), GFP_KERNEL);
+       if (!csi2)
+               return AE_OK;
+
+       csi2->handle = handle;
+       list_replace(&inst_context.res_list, &csi2->buses);

And here, csi2 represents a "sender" device along with a list of "receiver" devices pointed to by its CSI-2 _CRS entries, right?

Quoted 25 lines
drivers/acpi/mipi.c new file +25 -0
+       list_add(&csi2->list, list);
+
+       return AE_OK;
+}
+
+struct acpi_handle_ref {
+       acpi_handle handle;
+       unsigned int count;
+};
+
+#define NO_CSI2_PORT (~1U)
+
+static int crs_handle_cmp(const void *__a, const void *__b)
+{
+       const struct acpi_handle_ref *a = __a, *b = __b;
+
+       return a->handle < b->handle ? -1 : a->handle > b->handle;
+}
+
+static void crs_csi2_release(struct list_head *crs_csi2_handles)
+{
+       struct crs_csi2 *csi2, *csi2_tmp;
+
+       list_for_each_entry_safe(csi2, csi2_tmp, crs_csi2_handles, list) {
+               struct crs_csi2_instance *inst, *inst_tmp;

drivers/acpi/mipi.c line 166

Quoted 10 lines
drivers/acpi/mipi.c new file +10 -0
+
+               list_for_each_entry_safe(inst, inst_tmp, &csi2->buses, list) {
+                       list_del(&inst->list);
+                       kfree(inst);
+               }
+
+               list_del(&csi2->list);
+               kfree(csi2);
+       }
+}

So far, not even a word of description of what's going on.

drivers/acpi/mipi.c line 171

Quoted 5 lines
drivers/acpi/mipi.c new file +5 -0
+
+/**
+ * acpi_bus_scan_crs_csi2 - Scan a device and its child devices for _CRS CSI-2
+ *
+ * @handle: ACPI handle to scan

"Root of the ACPI Namespace branch to scan"

drivers/acpi/mipi.c line 176

Quoted 5 lines
drivers/acpi/mipi.c new file +5 -0
+ *
+ * This function does a number of things:
+ *
+ * 1. Scan an ACPI device and its children for _CRS CSI-2 instances. The
+ *    instances are stored for later use.

This is not just the children, but everything under @handle, because the callback doesn't terminate the walk.

And if @handle happens to be the ACPI Namespace root, the function will scan all of it AFAICS.

drivers/acpi/mipi.c line 187

Quoted 11 lines
drivers/acpi/mipi.c new file +11 -0
+ *
+ * 2. Count how many references to other devices _CRS CSI-2 instances have in
+ *    total.
+ *
+ * 3. Count the number of references to other devices for each _CRS CSI-2
+ *    instance.
+ *
+ * 4. Allocate memory for swnodes each ACPI device requires later on, and
+ *    generate a list of such allocations.
+ *
+ * Note that struct acpi_device isn't available yet at this time.

Why exactly can't this be done after the struct acpi_device objects have been created, but before binding any drivers etc to them?

Quoted 5 lines
drivers/acpi/mipi.c new file +5 -0
+ *
+ * acpi_scan_lock in scan.c must be held when calling this function.
+ */
+void acpi_bus_scan_crs_csi2(acpi_handle handle)
+{

drivers/acpi/mipi.c line 202

Quoted 10 lines
drivers/acpi/mipi.c new file +10 -0
+       struct acpi_handle_ref *handle_refs;
+       struct acpi_handle_ref *this = NULL;
+       LIST_HEAD(crs_csi2_handles);
+       unsigned int handle_count = 0, this_count;
+       unsigned int curr = 0;
+       struct crs_csi2 *csi2;
+
+       /* Collect the devices that have a _CRS CSI-2 resource */
+       acpi_walk_namespace(ACPI_TYPE_DEVICE, handle, ACPI_UINT32_MAX,
+                           scan_check_crs_csi2, NULL, &crs_csi2_handles, NULL);

So the result of this will be a list of "senders" where each entry represents one device along with a list of remote "receivers" connected to it. Is this correct?

drivers/acpi/mipi.c line 215

Quoted 13 lines
drivers/acpi/mipi.c new file +13 -0
+
+       /*
+        * Figure out how much temporary storage we need for counting
+        * connections in each device.
+        */
+       list_for_each_entry(csi2, &crs_csi2_handles, list) {
+               struct crs_csi2_instance *inst;
+
+               handle_count++;
+
+               list_for_each_entry(inst, &csi2->buses, list)
+                       handle_count++;
+       }

I would suggest counting stuff while creating the data structure, so this extra walk would not be necessary.

Quoted 20 lines
drivers/acpi/mipi.c new file +20 -0
+
+       /* No handles? Bail out here. */
+       if (!handle_count)
+               return;
+
+       handle_refs = kcalloc(handle_count + 1, sizeof(*handle_refs),
+                             GFP_KERNEL);
+       if (!handle_refs) {
+               acpi_handle_debug(handle, "no memory for %u handle refs\n",
+                                 handle_count + 1);
+               return;
+       }
+
+       /* Associate handles to the number of references. */
+       list_for_each_entry(csi2, &crs_csi2_handles, list) {
+               struct crs_csi2_instance *inst;
+               struct acpi_handle_ref *handle_ref;
+
+               handle_ref = &handle_refs[curr++];
+               handle_ref->handle = csi2->handle;

drivers/acpi/mipi.c line 245

Quoted 10 lines
drivers/acpi/mipi.c new file +10 -0
+
+               list_for_each_entry(inst, &csi2->buses, list) {
+                       handle_refs[curr].handle = inst->remote_handle;
+                       handle_refs[curr].count = 1;
+                       handle_ref->count++;
+                       curr++;
+               }
+       }
+
+       handle_refs[curr].handle = NULL;

So the above basically creates a flat array of entries that each contain an ACPI handle of (an ACPI object representing) a CSI-2 device and the number of other CSI-2 devices that it can talk to.

But one device can appear in multiple entries, which is why it is sorted below and then the entries corresponding to one device are aggregated.

Quoted 24 lines
drivers/acpi/mipi.c new file +24 -0
+
+       sort(handle_refs, handle_count, sizeof(*handle_refs), crs_handle_cmp,
+            NULL);
+
+       /*
+        * Finally count references in each handle, allocate space for device
+        * specific ports, properties and fill the _CRS CSI2 descriptor
+        * originated data.
+        */
+       this = handle_refs;
+       this_count = this->count;
+       for (curr = 1; curr < handle_count + 1; curr++) {
+               struct acpi_device_software_nodes *ads;
+               struct crs_csi2_swnodes *swnodes;
+               size_t alloc_size;
+               void *end;
+
+               if (this->handle == handle_refs[curr].handle) {
+                       this_count += handle_refs[curr].count;
+                       continue;
+               }
+
+               /*
+                * Allocate memory for ports, node pointers (number of nodes +

drivers/acpi/mipi.c line 279

Quoted 10 lines
drivers/acpi/mipi.c new file +10 -0
+                * 1 (guardian), nodes (root + number of ports * 2 (for for
+                * every port there is an endpoint)).
+                */
+               if (check_mul_overflow(sizeof(*ads->ports) +
+                                      sizeof(*ads->nodes) * 2 +
+                                      sizeof(*ads->nodeptrs) * 2,
+                                      (size_t)this_count, &alloc_size) ||
+                   check_add_overflow(sizeof(*ads) + sizeof(*ads->nodes) +
+                                      sizeof(*ads->nodeptrs) * 2,
+                                      alloc_size, &alloc_size)) {

Please try to split this conditional into smaller pieces. As is, it is almost completely unreadable.

drivers/acpi/mipi.c line 281

Quoted 2 lines
drivers/acpi/mipi.c new file +2 -0
+                       acpi_handle_warn(handle, "too many handles (%u)",
+                                        this_count);

And the warning is utterly useless except for debugging.

drivers/acpi/mipi.c line 293

Quoted 12 lines
drivers/acpi/mipi.c new file +12 -0
+                       continue;
+               }
+
+               swnodes = kzalloc(sizeof(*swnodes), GFP_KERNEL);
+               ads = kzalloc(alloc_size, GFP_KERNEL);
+               ads->ports = (void *)(ads + 1);
+               ads->nodes = (void *)(ads->ports + this_count);
+               ads->nodeptrs = (void *)(ads->nodes +
+                                        this_count * 2 + 1);
+               end = ads->nodeptrs + this_count * 2 + 2;
+               if (!swnodes || !ads ||
+                   WARN_ON((void *)ads + alloc_size != end)) {

Why WARN_ON()? Is this so important?

drivers/acpi/mipi.c line 297

Quoted 4 lines
drivers/acpi/mipi.c new file +4 -0
+                       kfree(swnodes);
+                       kfree(ads);
+                       acpi_handle_debug(handle,
+                                         "cannot allocate for %u swnodes\n",

Say "software nodes" in the message and please mention CSI-2 in it too.

Quoted 4 lines
drivers/acpi/mipi.c new file +4 -0
+                                         this_count);
+               } else {
+                       unsigned int i;
+

drivers/acpi/mipi.c line 311

Quoted 10 lines
drivers/acpi/mipi.c new file +10 -0
+                       ads->num_ports = this_count;
+                       for (i = 0; i < this_count * 2 + 1; i++)
+                               ads->nodeptrs[i] = &ads->nodes[i];
+                       ads->nodeptrs[i] = NULL;
+                       for (i = 0; i < this_count; i++)
+                               ads->ports[i].port_nr = NO_CSI2_PORT;
+                       swnodes->handle = this->handle;
+                       swnodes->ads = ads;
+                       list_add(&swnodes->list, &crs_csi2_swnodes);
+               }

The allocation part above can be done in a separate function with a description of what's going on.

It is really hard to figure out what happens here from the code and it shouldn't even be necessary. Please explain what you are doing.

Quoted 14 lines
drivers/acpi/mipi.c new file +9 -0
+
+               this = &handle_refs[curr];
+               this_count = this->count;
+       }
+
+       kfree(handle_refs);
+
+       crs_csi2_release(&crs_csi2_handles);
+}
drivers/acpi/scan.c +0 -0
index 0c6f06abe3f47..50de874b8f208 100644

drivers/acpi/scan.c EXPORT_SYMBOL_GPL(acpi_dev_get_next_consumer_dev); · line 2431

Quoted 9 lines
drivers/acpi/scan.c +3 -0
@@ -2423,9 +2423,12 @@ EXPORT_SYMBOL_GPL(acpi_dev_get_next_consumer_dev);
 int acpi_bus_scan(acpi_handle handle)
 {
        struct acpi_device *device = NULL;
+       int ret = 0;
 
        acpi_bus_scan_second_pass = false;
 
+       acpi_bus_scan_crs_csi2(handle);
+

This effectively adds an extra walk over the entire ACPI Namespace which is pure useless overhead on systems without CSI-2 descriptors.

I'm predicting that server people who care about boot times won't like it.

Quoted 42 lines
drivers/acpi/scan.c +9 -4
        /* Pass 1: Avoid enumerating devices with missing dependencies. */
 
        if (ACPI_SUCCESS(acpi_bus_check_add(handle, true, &device)))
@@ -2433,13 +2436,15 @@ int acpi_bus_scan(acpi_handle handle)
                                    acpi_bus_check_add_1, NULL, NULL,
                                    (void **)&device);
 
-       if (!device)
-               return -ENODEV;
+       if (!device) {
+               ret = -ENODEV;
+               goto out_release;
+       }
 
        acpi_bus_attach(device, (void *)true);
 
        if (!acpi_bus_scan_second_pass)
-               return 0;
+               goto out_release;
 
        /* Pass 2: Enumerate all of the remaining devices. */
 
@@ -2452,7 +2457,10 @@ int acpi_bus_scan(acpi_handle handle)
 
        acpi_bus_attach(device, NULL);
 
-       return 0;
+out_release:
+       acpi_crs_csi2_swnodes_del_free();
+
+       return ret;
 }
 EXPORT_SYMBOL(acpi_bus_scan);
 
include/acpi/acpi_bus.h +0 -0
index e44be31115a67..a05fe22c1175c 100644
@@ -360,6 +360,17 @@ struct acpi_device_data {
 
 struct acpi_gpio_mapping;
 

include/acpi/acpi_bus.h struct acpi_device_data { · line 372

Quoted 10 lines
include/acpi/acpi_bus.h +10 -0
+struct acpi_device_software_node_port {
+       unsigned int port_nr;
+};
+
+struct acpi_device_software_nodes {
+       struct acpi_device_software_node_port *ports;
+       struct software_node *nodes;
+       const struct software_node **nodeptrs;
+       unsigned int num_ports;
+};

No description?

Quoted 5 lines
include/acpi/acpi_bus.h +1 -1
+
 /* Device */
 struct acpi_device {
        u32 pld_crc;
--
author reviewer
#11 lore raw

Re: [PATCH v3 2/8] ACPI: property: Parse _CRS CSI-2 descriptor

Hi Rafael,

Thank you for the review.

Quoted 104 lines · Rafael J. Wysocki
Quoted 102 lines · Sakari Ailus

Parse newly added ACPI _CRS CSI-2 descriptor for CSI-2 and camera configuration. For now, only figure out where the descriptor is present in order to allow adding information from it to related devices.

---
 drivers/acpi/Makefile   |   2 +-
 drivers/acpi/internal.h |   7 +
 drivers/acpi/mipi.c     | 320 ++++++++++++++++++++++++++++++++++++++++
 drivers/acpi/scan.c     |  16 +-
 include/acpi/acpi_bus.h |  11 ++
 5 files changed, 351 insertions(+), 5 deletions(-)
 create mode 100644 drivers/acpi/mipi.c
drivers/acpi/Makefile +1 -1
index a5b649e71ab1b..a98fa1bc15548 100644
@@ -37,7 +37,7 @@ acpi-$(CONFIG_ACPI_SLEEP)     += proc.o
 # ACPI Bus and Device Drivers
 #
 acpi-y                         += bus.o glue.o
-acpi-y                         += scan.o
+acpi-y                         += scan.o mipi.o
 acpi-y                         += resource.o
 acpi-y                         += acpi_processor.o
 acpi-y                         += processor_core.o
drivers/acpi/internal.h +7 -0
index ec584442fb298..1ec4aa92bf17a 100644
@@ -282,4 +282,11 @@ void acpi_init_lpit(void);
 static inline void acpi_init_lpit(void) { }
 #endif
 
+/*--------------------------------------------------------------------------
+                               ACPI and MIPI DisCo for Imaging conversion
+  -------------------------------------------------------------------------- */
+
+void acpi_crs_csi2_swnodes_del_free(void);
+void acpi_bus_scan_crs_csi2(acpi_handle handle);
+
 #endif /* _ACPI_INTERNAL_H_ */
drivers/acpi/mipi.c new file +52 -0
new file mode 100644
index 0000000000000..269a6965d56e6
@@ -0,0 +1,320 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * MIPI DisCo for Imaging support.
+ *
+ * Copyright (C) 2023 Intel Corporation
+ */
+
+#include <linux/acpi.h>
+#include <linux/list.h>
+#include <linux/module.h>
+#include <linux/overflow.h>
+#include <linux/types.h>
+#include <linux/slab.h>
+#include <linux/sort.h>
+#include <linux/string.h>
+
+#include "internal.h"
+
+struct crs_csi2_swnodes {
+       struct list_head list;
+       acpi_handle handle;
+       struct acpi_device_software_nodes *ads;
+};
+
+static LIST_HEAD(crs_csi2_swnodes);
+
+static void crs_csi2_swnode_del_free(struct crs_csi2_swnodes *swnodes)
+{
+       list_del(&swnodes->list);
+       kfree(swnodes);
+}
+
+void acpi_crs_csi2_swnodes_del_free(void)
+{
+       struct crs_csi2_swnodes *swnodes, *swnodes_tmp;
+
+       list_for_each_entry_safe(swnodes, swnodes_tmp, &crs_csi2_swnodes, list)
+               crs_csi2_swnode_del_free(swnodes);
+}
+
+struct crs_csi2_instance {
+       struct list_head list;
+       struct acpi_resource_csi2_serialbus csi2;
+       acpi_handle remote_handle;
+       char remote_name[];
+};
+
+struct crs_csi2 {
+       struct list_head list;
+       acpi_handle handle;
+       struct list_head buses;
+};
Quoted 8 lines

drivers/acpi/mipi.c line 57

Quoted 5 lines
drivers/acpi/mipi.c new file +5 -0
+
+struct scan_check_crs_csi2_context {
+       struct list_head res_list;
+       acpi_handle handle;
+};

All of the structures above need to be described, including what they represent and what they are for.

I'll add comments explaining what these are for.

Quoted 28 lines
Quoted 28 lines
drivers/acpi/mipi.c new file +27 -0
+
+static acpi_status scan_check_crs_csi2_instance(struct acpi_resource *res,
+                                               void *context)
+{
+       struct scan_check_crs_csi2_context *inst_context = context;
+       struct acpi_resource_csi2_serialbus *csi2;
+       struct crs_csi2_instance *inst;
+       acpi_handle remote_handle;
+       acpi_status status;
+
+       if (res->type != ACPI_RESOURCE_TYPE_SERIAL_BUS)
+               return AE_OK;
+
+       csi2 = &res->data.csi2_serial_bus;
+
+       if (csi2->type != ACPI_RESOURCE_SERIAL_TYPE_CSI2)
+               return AE_OK;
+
+       if (!csi2->resource_source.string_length) {
+               acpi_handle_debug(inst_context->handle,
+                                 "invalid resource source string length\n");
+               return AE_OK;
+       }
+
+       status = acpi_get_handle(NULL, csi2->resource_source.string_ptr,
+                                &remote_handle);
+       if (status != AE_OK) {
Quoted 8 lines

drivers/acpi/mipi.c line 89

Quoted 5 lines
drivers/acpi/mipi.c new file +5 -0
+               acpi_handle_warn(inst_context->handle,
+                                "cannot get handle for %s\n",
+                                csi2->resource_source.string_ptr);
+               return AE_OK;
+       }

After this, remote_handle will be the ACPI handle of an ACPI object representing a remote CSI-2 device IIUC.

Yes.

Quoted 5 lines

drivers/acpi/mipi.c line 91

Quoted 3 lines
drivers/acpi/mipi.c new file +2 -0
+
+       inst = kmalloc(struct_size(inst, remote_name,

This line break is confusing.

I can unwrap it, resulting a rather long line, but I guess that's not an issue.

Quoted 11 lines

drivers/acpi/mipi.c line 97

Quoted 7 lines
drivers/acpi/mipi.c new file +6 -0
+                                  csi2->resource_source.string_length),
+                      GFP_KERNEL);
+       if (!inst)
+               return AE_OK;
+
+       inst->csi2 = *csi2;

A pointer to the entire csi2 structure is stored in ints, so why is it still necessary to copy remote_name and remote_handle extracted from it to the additional fields?

csi2->resource_source.string_ptr is a pointer pointing somewhere else (DSDT binary I suppose?) and we're making a local copy of that here. The purpose is to not rely on this memory being available later on.

Quoted 10 lines

drivers/acpi/mipi.c line 103

Quoted 7 lines
drivers/acpi/mipi.c new file +6 -0
+       strscpy(inst->remote_name, csi2->resource_source.string_ptr,
+               csi2->resource_source.string_length);
+       inst->csi2.resource_source.string_ptr = inst->remote_name;
+       inst->remote_handle = remote_handle;
+
+       list_add(&inst->list, &inst_context->res_list);

So this adds an entry to a list of "receivers" pointed to by the _CRS for one "sender"?

Correct. (Usually "transmitter" is used in this context, but same thing.)

Quoted 18 lines
Quoted 18 lines
drivers/acpi/mipi.c new file +17 -0
+
+       return AE_OK;
+}
+
+static acpi_status scan_check_crs_csi2(acpi_handle handle, u32 nesting_level,
+                                      void *context, void **ret)
+{
+       struct scan_check_crs_csi2_context inst_context = {
+               .handle = handle,
+               .res_list = LIST_HEAD_INIT(inst_context.res_list),
+       };
+       struct list_head *list = context;
+       struct crs_csi2 *csi2;
+
+       INIT_LIST_HEAD(&inst_context.res_list);
+
+       acpi_walk_resources(handle, METHOD_NAME__CRS,
Quoted 8 lines

drivers/acpi/mipi.c line 124

Quoted 4 lines
drivers/acpi/mipi.c new file +4 -0
+                           scan_check_crs_csi2_instance, &inst_context);
+
+       if (list_empty(&inst_context.res_list))
+               return AE_OK;

At this point inst_context.res_list is a list of remote "receivers" pointed to by all of the CSI-2 _CRS entries for the device represented by @handle?

Yes.

Quoted 11 lines

drivers/acpi/mipi.c line 131

Quoted 8 lines
drivers/acpi/mipi.c new file +7 -0
+
+       csi2 = kmalloc(sizeof(*csi2), GFP_KERNEL);
+       if (!csi2)
+               return AE_OK;
+
+       csi2->handle = handle;
+       list_replace(&inst_context.res_list, &csi2->buses);

And here, csi2 represents a "sender" device along with a list of "receiver" devices pointed to by its CSI-2 _CRS entries, right?

Yes. I can add comments detailing this.

Quoted 39 lines
Quoted 26 lines
drivers/acpi/mipi.c new file +25 -0
+       list_add(&csi2->list, list);
+
+       return AE_OK;
+}
+
+struct acpi_handle_ref {
+       acpi_handle handle;
+       unsigned int count;
+};
+
+#define NO_CSI2_PORT (~1U)
+
+static int crs_handle_cmp(const void *__a, const void *__b)
+{
+       const struct acpi_handle_ref *a = __a, *b = __b;
+
+       return a->handle < b->handle ? -1 : a->handle > b->handle;
+}
+
+static void crs_csi2_release(struct list_head *crs_csi2_handles)
+{
+       struct crs_csi2 *csi2, *csi2_tmp;
+
+       list_for_each_entry_safe(csi2, csi2_tmp, crs_csi2_handles, list) {
+               struct crs_csi2_instance *inst, *inst_tmp;

drivers/acpi/mipi.c line 166

Quoted 10 lines
drivers/acpi/mipi.c new file +10 -0
+
+               list_for_each_entry_safe(inst, inst_tmp, &csi2->buses, list) {
+                       list_del(&inst->list);
+                       kfree(inst);
+               }
+
+               list_del(&csi2->list);
+               kfree(csi2);
+       }
+}

So far, not even a word of description of what's going on.

Quoted 7 lines

drivers/acpi/mipi.c line 171

Quoted 5 lines
drivers/acpi/mipi.c new file +5 -0
+
+/**
+ * acpi_bus_scan_crs_csi2 - Scan a device and its child devices for _CRS CSI-2
+ *
+ * @handle: ACPI handle to scan

"Root of the ACPI Namespace branch to scan"

Yes...

Quoted 7 lines

drivers/acpi/mipi.c line 176

Quoted 6 lines
drivers/acpi/mipi.c new file +5 -0
+ *
+ * This function does a number of things:
+ *
+ * 1. Scan an ACPI device and its children for _CRS CSI-2 instances. The
+ *    instances are stored for later use.
Quoted 5 lines

This is not just the children, but everything under @handle, because the callback doesn't terminate the walk.

And if @handle happens to be the ACPI Namespace root, the function will scan all of it AFAICS.

Correct. I'll change this to convey the traversal isn't about immediate children only.

Quoted 13 lines

drivers/acpi/mipi.c line 187

Quoted 12 lines
drivers/acpi/mipi.c new file +11 -0
+ *
+ * 2. Count how many references to other devices _CRS CSI-2 instances have in
+ *    total.
+ *
+ * 3. Count the number of references to other devices for each _CRS CSI-2
+ *    instance.
+ *
+ * 4. Allocate memory for swnodes each ACPI device requires later on, and
+ *    generate a list of such allocations.
+ *
+ * Note that struct acpi_device isn't available yet at this time.
Quoted 2 lines

Why exactly can't this be done after the struct acpi_device objects have been created, but before binding any drivers etc to them?

It could be done, yes, but you'd have to create all acpi_device objects before any of them may get probed. The data is gathered one _CRS CSI2 descriptor at a time and all remote devices aren't knon before all have been parsed.

Quoted 12 lines
Quoted 12 lines
drivers/acpi/mipi.c new file +11 -0
+ *
+ * acpi_scan_lock in scan.c must be held when calling this function.
+ */
+void acpi_bus_scan_crs_csi2(acpi_handle handle)
+{
+       struct acpi_handle_ref *handle_refs;
+       struct acpi_handle_ref *this = NULL;
+       LIST_HEAD(crs_csi2_handles);
+       unsigned int handle_count = 0, this_count;
+       unsigned int curr = 0;
+       struct crs_csi2 *csi2;
Quoted 8 lines

drivers/acpi/mipi.c line 202

Quoted 4 lines
drivers/acpi/mipi.c new file +4 -0
+
+       /* Collect the devices that have a _CRS CSI-2 resource */
+       acpi_walk_namespace(ACPI_TYPE_DEVICE, handle, ACPI_UINT32_MAX,
+                           scan_check_crs_csi2, NULL, &crs_csi2_handles, NULL);

So the result of this will be a list of "senders" where each entry represents one device along with a list of remote "receivers" connected to it. Is this correct?

Correct.

Quoted 17 lines
Quoted 4 lines
drivers/acpi/mipi.c new file +3 -0
+
+       /*
+        * Figure out how much temporary storage we need for counting

drivers/acpi/mipi.c line 215

Quoted 10 lines
drivers/acpi/mipi.c new file +10 -0
+        * connections in each device.
+        */
+       list_for_each_entry(csi2, &crs_csi2_handles, list) {
+               struct crs_csi2_instance *inst;
+
+               handle_count++;
+
+               list_for_each_entry(inst, &csi2->buses, list)
+                       handle_count++;
+       }

I would suggest counting stuff while creating the data structure, so this extra walk would not be necessary.

Note that the number of entries in these lists is typically small, no more than 20 in total, so this isn't really taking much time. But I believe the counting can be done equally sell elsewhere, too, if you prefer that.

Quoted 36 lines
Quoted 21 lines
drivers/acpi/mipi.c new file +20 -0
+
+       /* No handles? Bail out here. */
+       if (!handle_count)
+               return;
+
+       handle_refs = kcalloc(handle_count + 1, sizeof(*handle_refs),
+                             GFP_KERNEL);
+       if (!handle_refs) {
+               acpi_handle_debug(handle, "no memory for %u handle refs\n",
+                                 handle_count + 1);
+               return;
+       }
+
+       /* Associate handles to the number of references. */
+       list_for_each_entry(csi2, &crs_csi2_handles, list) {
+               struct crs_csi2_instance *inst;
+               struct acpi_handle_ref *handle_ref;
+
+               handle_ref = &handle_refs[curr++];
+               handle_ref->handle = csi2->handle;

drivers/acpi/mipi.c line 245

Quoted 10 lines
drivers/acpi/mipi.c new file +10 -0
+
+               list_for_each_entry(inst, &csi2->buses, list) {
+                       handle_refs[curr].handle = inst->remote_handle;
+                       handle_refs[curr].count = 1;
+                       handle_ref->count++;
+                       curr++;
+               }
+       }
+
+       handle_refs[curr].handle = NULL;

So the above basically creates a flat array of entries that each contain an ACPI handle of (an ACPI object representing) a CSI-2 device and the number of other CSI-2 devices that it can talk to.

Quoted 3 lines

But one device can appear in multiple entries, which is why it is sorted below and then the entries corresponding to one device are aggregated.

Correct.

Quoted 30 lines
Quoted 30 lines
drivers/acpi/mipi.c new file +29 -0
+
+       sort(handle_refs, handle_count, sizeof(*handle_refs), crs_handle_cmp,
+            NULL);
+
+       /*
+        * Finally count references in each handle, allocate space for device
+        * specific ports, properties and fill the _CRS CSI2 descriptor
+        * originated data.
+        */
+       this = handle_refs;
+       this_count = this->count;
+       for (curr = 1; curr < handle_count + 1; curr++) {
+               struct acpi_device_software_nodes *ads;
+               struct crs_csi2_swnodes *swnodes;
+               size_t alloc_size;
+               void *end;
+
+               if (this->handle == handle_refs[curr].handle) {
+                       this_count += handle_refs[curr].count;
+                       continue;
+               }
+
+               /*
+                * Allocate memory for ports, node pointers (number of nodes +
+                * 1 (guardian), nodes (root + number of ports * 2 (for for
+                * every port there is an endpoint)).
+                */
+               if (check_mul_overflow(sizeof(*ads->ports) +
+                                      sizeof(*ads->nodes) * 2 +
Quoted 8 lines

drivers/acpi/mipi.c line 279

Quoted 5 lines
drivers/acpi/mipi.c new file +5 -0
+                                      sizeof(*ads->nodeptrs) * 2,
+                                      (size_t)this_count, &alloc_size) ||
+                   check_add_overflow(sizeof(*ads) + sizeof(*ads->nodes) +
+                                      sizeof(*ads->nodeptrs) * 2,
+                                      alloc_size, &alloc_size)) {

Please try to split this conditional into smaller pieces. As is, it is almost completely unreadable.

Yes.

Quoted 5 lines

drivers/acpi/mipi.c line 281

Quoted 3 lines
drivers/acpi/mipi.c new file +2 -0
+                       acpi_handle_warn(handle, "too many handles (%u)",
+                                        this_count);

And the warning is utterly useless except for debugging.

It is there exactly for that purpose. Although this isn't a serious matter as such, it is very likely to signal of a bug (most likely in the code).

Quoted 14 lines

drivers/acpi/mipi.c line 293

Quoted 13 lines
drivers/acpi/mipi.c new file +12 -0
+                       continue;
+               }
+
+               swnodes = kzalloc(sizeof(*swnodes), GFP_KERNEL);
+               ads = kzalloc(alloc_size, GFP_KERNEL);
+               ads->ports = (void *)(ads + 1);
+               ads->nodes = (void *)(ads->ports + this_count);
+               ads->nodeptrs = (void *)(ads->nodes +
+                                        this_count * 2 + 1);
+               end = ads->nodeptrs + this_count * 2 + 2;
+               if (!swnodes || !ads ||
+                   WARN_ON((void *)ads + alloc_size != end)) {
Quoted 1 line

Why WARN_ON()? Is this so important?

Sanity check. There are two allocations there with a different lifetime, and the size of the other one is somewhat complicated to calculate.

Quoted 7 lines

drivers/acpi/mipi.c line 297

Quoted 5 lines
drivers/acpi/mipi.c new file +4 -0
+                       kfree(swnodes);
+                       kfree(ads);
+                       acpi_handle_debug(handle,
+                                         "cannot allocate for %u swnodes\n",

Say "software nodes" in the message and please mention CSI-2 in it too.

Yes...

Quoted 43 lines
Quoted 5 lines
drivers/acpi/mipi.c new file +4 -0
+                                         this_count);
+               } else {
+                       unsigned int i;
+

drivers/acpi/mipi.c line 311

Quoted 10 lines
drivers/acpi/mipi.c new file +10 -0
+                       ads->num_ports = this_count;
+                       for (i = 0; i < this_count * 2 + 1; i++)
+                               ads->nodeptrs[i] = &ads->nodes[i];
+                       ads->nodeptrs[i] = NULL;
+                       for (i = 0; i < this_count; i++)
+                               ads->ports[i].port_nr = NO_CSI2_PORT;
+                       swnodes->handle = this->handle;
+                       swnodes->ads = ads;
+                       list_add(&swnodes->list, &crs_csi2_swnodes);
+               }

The allocation part above can be done in a separate function with a description of what's going on.

It is really hard to figure out what happens here from the code and it shouldn't even be necessary. Please explain what you are doing.

Quoted 21 lines
drivers/acpi/mipi.c new file +9 -0
+
+               this = &handle_refs[curr];
+               this_count = this->count;
+       }
+
+       kfree(handle_refs);
+
+       crs_csi2_release(&crs_csi2_handles);
+}
drivers/acpi/scan.c +1 -0
index 0c6f06abe3f47..50de874b8f208 100644
@@ -2423,9 +2423,12 @@ EXPORT_SYMBOL_GPL(acpi_dev_get_next_consumer_dev);
 int acpi_bus_scan(acpi_handle handle)
 {
        struct acpi_device *device = NULL;
+       int ret = 0;
 
        acpi_bus_scan_second_pass = false;
 
Quoted 7 lines

drivers/acpi/scan.c EXPORT_SYMBOL_GPL(acpi_dev_get_next_consumer_dev); · line 2431

Quoted 2 lines
drivers/acpi/scan.c +2 -0
+       acpi_bus_scan_crs_csi2(handle);
+

This effectively adds an extra walk over the entire ACPI Namespace which is pure useless overhead on systems without CSI-2 descriptors.

I'm predicting that server people who care about boot times won't like it.

I'm certainly fine with doing this with fewer walks, but it requires parsing the _CRS objects (as earlier in the patch) during the same traversal than devices are created. But any driver probing must be postponed after this traversal so we have found all the devices with CSI-2 connections (and created software nodes for them).

This is so by design, only one end of a connection has any information about it.

Quoted 43 lines
Quoted 43 lines
drivers/acpi/scan.c +9 -4
        /* Pass 1: Avoid enumerating devices with missing dependencies. */
 
        if (ACPI_SUCCESS(acpi_bus_check_add(handle, true, &device)))
@@ -2433,13 +2436,15 @@ int acpi_bus_scan(acpi_handle handle)
                                    acpi_bus_check_add_1, NULL, NULL,
                                    (void **)&device);
 
-       if (!device)
-               return -ENODEV;
+       if (!device) {
+               ret = -ENODEV;
+               goto out_release;
+       }
 
        acpi_bus_attach(device, (void *)true);
 
        if (!acpi_bus_scan_second_pass)
-               return 0;
+               goto out_release;
 
        /* Pass 2: Enumerate all of the remaining devices. */
 
@@ -2452,7 +2457,10 @@ int acpi_bus_scan(acpi_handle handle)
 
        acpi_bus_attach(device, NULL);
 
-       return 0;
+out_release:
+       acpi_crs_csi2_swnodes_del_free();
+
+       return ret;
 }
 EXPORT_SYMBOL(acpi_bus_scan);
 
include/acpi/acpi_bus.h +0 -0
index e44be31115a67..a05fe22c1175c 100644
@@ -360,6 +360,17 @@ struct acpi_device_data {
 
 struct acpi_gpio_mapping;
 
Quoted 12 lines

include/acpi/acpi_bus.h struct acpi_device_data { · line 372

Quoted 10 lines
include/acpi/acpi_bus.h +10 -0
+struct acpi_device_software_node_port {
+       unsigned int port_nr;
+};
+
+struct acpi_device_software_nodes {
+       struct acpi_device_software_node_port *ports;
+       struct software_node *nodes;
+       const struct software_node **nodeptrs;
+       unsigned int num_ports;
+};

No description?

None of the structs in that file (or generally elsewhere in vicinity) have any helpful comments so I was just following a pattern. :-) But perhaps these would deserve some. I can add them here, and see if other places could benefit from them as well (plus your comments earlier in the patch).

Quoted 6 lines
Quoted 6 lines
include/acpi/acpi_bus.h +1 -1
+
 /* Device */
 struct acpi_device {
        u32 pld_crc;
--
Signature · 3 lines
Kind regards,

Sakari Ailus
author reviewer
#12 lore raw

PATCH v3 3/8 device property: Add SOFTWARE_NODE() macro for defining software nodes

In reply to Sakari Ailus

Add SOFTWARE_NODE() macro in order to make defining software nodes look nicer. This is analogous to different PROPERTY_ENTRY_*() macros for defining properties.

---
 include/linux/property.h | 7 +++++++
 1 file changed, 7 insertions(+)
include/linux/property.h +7 -0
index 37179e3abad5c..6745a86bc9b97 100644
@@ -477,6 +477,13 @@ struct software_node {
 	const struct property_entry *properties;
 };
 
+#define SOFTWARE_NODE(_name_, _properties_, _parent_)	\
+	(struct software_node) {			\
+		.name = _name_,				\
+		.properties = _properties_,		\
+		.parent = _parent_,			\
+	}
+
 bool is_software_node(const struct fwnode_handle *fwnode);
 const struct software_node *
 to_software_node(const struct fwnode_handle *fwnode);
Signature · 1 line
2.30.2
author reviewer
#13 lore raw

PATCH v3 4/8 ACPI: property: Generate camera swnodes for ACPI and DisCo for Imaging

In reply to Sakari Ailus

Generate software nodes for information parsed from ACPI _CRS for CSI-2 as well as MIPI DisCo for Imaging spec. The software nodes are compliant with existing ACPI or DT definitions and are parsed by relevant drivers without changes.

---
 drivers/acpi/internal.h |   1 +
 drivers/acpi/mipi.c     | 353 ++++++++++++++++++++++++++++++++++++++++
 drivers/acpi/scan.c     |  19 +++
 include/acpi/acpi_bus.h |  49 ++++++
 4 files changed, 422 insertions(+)
drivers/acpi/internal.h +1 -0
index 1ec4aa92bf17a..fac87404e294c 100644
@@ -288,5 +288,6 @@ static inline void acpi_init_lpit(void) { }
 
 void acpi_crs_csi2_swnodes_del_free(void);
 void acpi_bus_scan_crs_csi2(acpi_handle handle);
+void acpi_init_swnodes(struct acpi_device *device);
 
 #endif /* _ACPI_INTERNAL_H_ */
drivers/acpi/mipi.c +353 -0

390 lines not on this page · raw message

drivers/acpi/scan.c +19 -0
index 50de874b8f208..9cc30c9c87080 100644
@@ -449,10 +449,28 @@ static void acpi_free_power_resources_lists(struct acpi_device *device)
 	}
 }
 
+static void acpi_free_swnodes(struct acpi_device *device)
+{
+	struct acpi_device_software_nodes *ads = device->swnodes;
+	struct fwnode_handle *primary;
+
+	if (!ads)
+		return;
+
+	software_node_unregister_node_group(ads->nodeptrs);
+	primary = acpi_fwnode_handle(device);
+	primary->secondary = NULL;
+	kfree(ads->nodes[ACPI_DEVICE_SWNODE_ROOT].name);
+	kfree(ads);
+
+	device->swnodes = NULL;
+}
+
 static void acpi_device_release(struct device *dev)
 {
 	struct acpi_device *acpi_dev = to_acpi_device(dev);
 
+	acpi_free_swnodes(acpi_dev);
 	acpi_free_properties(acpi_dev);
 	acpi_free_pnp_ids(&acpi_dev->pnp);
 	acpi_free_power_resources_lists(acpi_dev);
@@ -1771,6 +1789,7 @@ void acpi_init_device_object(struct acpi_device *device, acpi_handle handle,
 	acpi_device_get_busid(device);
 	acpi_set_pnp_ids(handle, &device->pnp, type);
 	acpi_init_properties(device);
+	acpi_init_swnodes(device);
 	acpi_bus_get_flags(device);
 	device->flags.match_driver = false;
 	device->flags.initialized = true;
include/acpi/acpi_bus.h +49 -0

69 lines not on this page · raw message

Signature · 1 line
2.30.2
#14 lore raw

Re: [PATCH v3 4/8] ACPI: property: Generate camera swnodes for ACPI and DisCo for Imaging

Quoted 4 lines · Sakari Ailus

Generate software nodes for information parsed from ACPI _CRS for CSI-2 as well as MIPI DisCo for Imaging spec. The software nodes are compliant with existing ACPI or DT definitions and are parsed by relevant drivers without changes.

...

Quoted 7 lines
drivers/acpi/mipi.c +7 -0
+static unsigned int next_csi2_port_index(struct acpi_device_software_nodes *ads,
+					 unsigned int port_nr)
+{
+	unsigned int i;
+
+	for (i = 0; i < ads->num_ports; i++) {
+		struct acpi_device_software_node_port *port = &ads->ports[i];

drivers/acpi/mipi.c struct acpi_handle_ref { · line 173

Quoted 10 lines
drivers/acpi/mipi.c +9 -0
+
+		if (port->port_nr == port_nr)
+			return i;
 
+		if (port->port_nr != NO_CSI2_PORT)
+			continue;
+
+		port->port_nr = port_nr;
+
+		return i;

Maybe it would be better to use the same pattern as above?

		if (port->port_nr == NO_CSI2_PORT) {
			port->port_nr = port_nr;
			return i;
		}
Quoted 4 lines
drivers/acpi/mipi.c +4 -0
+	}
+
+	return NO_CSI2_PORT;
+}

...

drivers/acpi/mipi.c void acpi_bus_scan_crs_csi2(acpi_handle handle) · line 465

Quoted 5 lines
drivers/acpi/mipi.c +5 -0
+static struct fwnode_handle *get_mipi_port_handle(struct acpi_device *device,
+						  unsigned int port)
+{
+	static const char mipi_port_prefix[] = "mipi-img-port-";
+	char mipi_port_name[sizeof(mipi_port_prefix) + 2];

I think the following will be better:

	char mipi_port_name[16];

drivers/acpi/mipi.c void acpi_bus_scan_crs_csi2(acpi_handle handle) · line 468

Quoted 2 lines
drivers/acpi/mipi.c +2 -0
+	if (snprintf(mipi_port_name, sizeof(mipi_port_name), "%s%u",
+		     mipi_port_prefix, port) >= sizeof(mipi_port_name)) {
	if (snprintf(mipi_port_name, sizeof(mipi_port_name), "mipi-img-port-%u",
		     port) >= sizeof(mipi_port_name)) {
Quoted 8 lines
drivers/acpi/mipi.c +8 -0
+		acpi_handle_info(acpi_device_handle(device),
+				 "mipi port name too long for port %u\n", port);
+		return NULL;
+	}
+
+	return fwnode_get_named_child_node(acpi_fwnode_handle(device),
+					   mipi_port_name);
+}

...

drivers/acpi/mipi.c void acpi_bus_scan_crs_csi2(acpi_handle handle) · line 488

Quoted 2 lines
drivers/acpi/mipi.c +2 -0
+	union {
+		u32 val;

OK. I stopped here, because I'm not sure you have really sent an updated version. For sure the val and union is not used.

drivers/acpi/mipi.c void acpi_bus_scan_crs_csi2(acpi_handle handle) · line 491

Quoted 3 lines
drivers/acpi/mipi.c +3 -0
+		/* Data lanes + the clock lane */
+		u8 val8[BITS_TO_BYTES(ARRAY_SIZE(port->data_lanes) + 1)];
+	} u;

...

Please, address all in v3.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#15 lore raw

Re: [PATCH v3 4/8] ACPI: property: Generate camera swnodes for ACPI and DisCo for Imaging

Quoted 1 line · Andy Shevchenko

On Thu, Jan 26, 2023 at 12:40:57AM +0200, Sakari Ailus wrote:

...

Quoted 1 line

Please, address all in v3.

I meant v4.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
author reviewer
#16 lore raw

Re: [PATCH v3 4/8] ACPI: property: Generate camera swnodes for ACPI and DisCo for Imaging

In reply to Andy Shevchenko

Hi Andy,

Quoted 28 lines · Andy Shevchenko
Quoted 4 lines · Sakari Ailus

Generate software nodes for information parsed from ACPI _CRS for CSI-2 as well as MIPI DisCo for Imaging spec. The software nodes are compliant with existing ACPI or DT definitions and are parsed by relevant drivers without changes.

...

Quoted 7 lines
drivers/acpi/mipi.c +7 -0
+static unsigned int next_csi2_port_index(struct acpi_device_software_nodes *ads,
+					 unsigned int port_nr)
+{
+	unsigned int i;
+
+	for (i = 0; i < ads->num_ports; i++) {
+		struct acpi_device_software_node_port *port = &ads->ports[i];

drivers/acpi/mipi.c struct acpi_handle_ref { · line 173

Quoted 10 lines
drivers/acpi/mipi.c +9 -0
+
+		if (port->port_nr == port_nr)
+			return i;
 
+		if (port->port_nr != NO_CSI2_PORT)
+			continue;
+
+		port->port_nr = port_nr;
+
+		return i;

Maybe it would be better to use the same pattern as above?

Quoted 4 lines
		if (port->port_nr == NO_CSI2_PORT) {
			port->port_nr = port_nr;
			return i;
		}

Works for me...

Quoted 16 lines
Quoted 5 lines
drivers/acpi/mipi.c +4 -0
+	}
+
+	return NO_CSI2_PORT;
+}

...

drivers/acpi/mipi.c void acpi_bus_scan_crs_csi2(acpi_handle handle) · line 465

Quoted 5 lines
drivers/acpi/mipi.c +5 -0
+static struct fwnode_handle *get_mipi_port_handle(struct acpi_device *device,
+						  unsigned int port)
+{
+	static const char mipi_port_prefix[] = "mipi-img-port-";
+	char mipi_port_name[sizeof(mipi_port_prefix) + 2];

I think the following will be better:

Quoted 1 line
	char mipi_port_name[16];

If the array is too short, this will generate a warning but... this is already handled better than that. Why not to keep it?

Quoted 19 lines

drivers/acpi/mipi.c void acpi_bus_scan_crs_csi2(acpi_handle handle) · line 468

Quoted 4 lines
drivers/acpi/mipi.c +2 -0
+	if (snprintf(mipi_port_name, sizeof(mipi_port_name), "%s%u",
+		     mipi_port_prefix, port) >= sizeof(mipi_port_name)) {
	if (snprintf(mipi_port_name, sizeof(mipi_port_name), "mipi-img-port-%u",
		     port) >= sizeof(mipi_port_name)) {
Quoted 8 lines
drivers/acpi/mipi.c +8 -0
+		acpi_handle_info(acpi_device_handle(device),
+				 "mipi port name too long for port %u\n", port);
+		return NULL;
+	}
+
+	return fwnode_get_named_child_node(acpi_fwnode_handle(device),
+					   mipi_port_name);
+}

...

Quoted 5 lines

drivers/acpi/mipi.c void acpi_bus_scan_crs_csi2(acpi_handle handle) · line 488

Quoted 2 lines
drivers/acpi/mipi.c +2 -0
+	union {
+		u32 val;

OK. I stopped here, because I'm not sure you have really sent an updated version. For sure the val and union is not used.

Indeed. My apologies --- I missed making a few changes I intended to do. I'll go through those once more and then send v4.

Quoted 4 lines
Quoted 4 lines
drivers/acpi/mipi.c +3 -0
+		/* Data lanes + the clock lane */
+		u8 val8[BITS_TO_BYTES(ARRAY_SIZE(port->data_lanes) + 1)];
+	} u;
Signature · 3 lines
Kind regards,

Sakari Ailus
#17 lore raw

Re: [PATCH v3 4/8] ACPI: property: Generate camera swnodes for ACPI and DisCo for Imaging

Quoted 2 lines · Sakari Ailus
Quoted 1 line · Andy Shevchenko

On Thu, Jan 26, 2023 at 12:40:57AM +0200, Sakari Ailus wrote:

...

Quoted 8 lines
Quoted 8 lines

drivers/acpi/mipi.c void acpi_bus_scan_crs_csi2(acpi_handle handle) · line 465

Quoted 5 lines
drivers/acpi/mipi.c +5 -0
+static struct fwnode_handle *get_mipi_port_handle(struct acpi_device *device,
+						  unsigned int port)
+{
+	static const char mipi_port_prefix[] = "mipi-img-port-";
+	char mipi_port_name[sizeof(mipi_port_prefix) + 2];

I think the following will be better:

Quoted 4 lines
Quoted 1 line
	char mipi_port_name[16];

If the array is too short, this will generate a warning but... this is already handled better than that. Why not to keep it?

We will get the format string self contained.

Quoted 14 lines
Quoted 14 lines

drivers/acpi/mipi.c void acpi_bus_scan_crs_csi2(acpi_handle handle) · line 468

Quoted 2 lines
drivers/acpi/mipi.c +2 -0
+	if (snprintf(mipi_port_name, sizeof(mipi_port_name), "%s%u",
+		     mipi_port_prefix, port) >= sizeof(mipi_port_name)) {
	if (snprintf(mipi_port_name, sizeof(mipi_port_name), "mipi-img-port-%u",
		     port) >= sizeof(mipi_port_name)) {
Quoted 8 lines
drivers/acpi/mipi.c +8 -0
+		acpi_handle_info(acpi_device_handle(device),
+				 "mipi port name too long for port %u\n", port);
+		return NULL;
+	}
+
+	return fwnode_get_named_child_node(acpi_fwnode_handle(device),
+					   mipi_port_name);
+}

...

Quoted 8 lines
Quoted 5 lines

drivers/acpi/mipi.c void acpi_bus_scan_crs_csi2(acpi_handle handle) · line 488

Quoted 2 lines
drivers/acpi/mipi.c +2 -0
+	union {
+		u32 val;

OK. I stopped here, because I'm not sure you have really sent an updated version. For sure the val and union is not used.

Indeed. My apologies --- I missed making a few changes I intended to do. I'll go through those once more and then send v4.

Thank you!

Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
drivers/acpi/mipi.c +3 -0
+		/* Data lanes + the clock lane */
+		u8 val8[BITS_TO_BYTES(ARRAY_SIZE(port->data_lanes) + 1)];
+	} u;
Signature · 2 lines
With Best Regards,
Andy Shevchenko
author reviewer
#18 lore raw

Re: [PATCH v3 4/8] ACPI: property: Generate camera swnodes for ACPI and DisCo for Imaging

Quoted 14 lines · Andy Shevchenko
Quoted 2 lines · Sakari Ailus
Quoted 1 line · Andy Shevchenko

On Thu, Jan 26, 2023 at 12:40:57AM +0200, Sakari Ailus wrote:

...

Quoted 8 lines
Quoted 8 lines

drivers/acpi/mipi.c void acpi_bus_scan_crs_csi2(acpi_handle handle) · line 465

Quoted 5 lines
drivers/acpi/mipi.c +5 -0
+static struct fwnode_handle *get_mipi_port_handle(struct acpi_device *device,
+						  unsigned int port)
+{
+	static const char mipi_port_prefix[] = "mipi-img-port-";
+	char mipi_port_name[sizeof(mipi_port_prefix) + 2];

I think the following will be better:

Quoted 6 lines
Quoted 4 lines
Quoted 1 line
	char mipi_port_name[16];

If the array is too short, this will generate a warning but... this is already handled better than that. Why not to keep it?

We will get the format string self contained.

I have earlier stated as my opinion that it is a lesser concern.

Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
drivers/acpi/mipi.c +2 -0
+	if (snprintf(mipi_port_name, sizeof(mipi_port_name), "%s%u",
+		     mipi_port_prefix, port) >= sizeof(mipi_port_name)) {
Signature · 1 line
Sakari Ailus
author reviewer
#19 lore raw

PATCH v3 5/8 ACPI: property: Dig "rotation" property for devices with CSI2 _CRS

In reply to Sakari Ailus

Dig "rotation" property value for devices with _CRS CSI2 resource descriptor. The value comes from _PLD (physical location of device) object, if it exists for the device.

This way camera sensor drivers that know the "rotation" property do not need to care about _PLD on ACPI.

---
 drivers/acpi/mipi.c     | 16 ++++++++++++++++
 include/acpi/acpi_bus.h |  1 +
 2 files changed, 17 insertions(+)
drivers/acpi/mipi.c +16 -0
index 5dbdf33a13831..b81bfa1edd8ad 100644
@@ -629,16 +629,32 @@ static void init_port_csi2_remote(struct acpi_device *device,
  */
 void acpi_init_swnodes(struct acpi_device *device)
 {
+	struct fwnode_handle *fwnode = acpi_fwnode_handle(device);
 	struct acpi_device_software_nodes *ads;
 	struct acpi_buffer buffer = { .length = ACPI_ALLOCATE_BUFFER };
 	struct fwnode_handle *primary;
+	struct acpi_pld_info *pld;
+	unsigned int prop_index = 0;
 	unsigned int i;
+	u32 val;
 	int ret;
 
 	device->swnodes = ads = crs_csi2_swnode_get(device->handle);
 	if (!ads)
 		return;
 
+	/*
+	 * Check if "rotation" property exists and if it doesn't but there's a
+	 * _PLD object, then get the rotation value from there.
+	 */
+	if (fwnode_property_read_u32(fwnode, "rotation", &val) &&
+	    ACPI_SUCCESS(acpi_get_physical_device_location(acpi_device_handle(device),
+							   &pld))) {
+		ads->dev_props[NEXT_PROPERTY(prop_index, DEV_ROTATION)] =
+			PROPERTY_ENTRY_U32("rotation", pld->rotation * 45U);
+		kfree(pld);
+	}
+
 	if (ACPI_FAILURE(acpi_get_name(device->handle, ACPI_FULL_PATHNAME, &buffer))) {
 		acpi_handle_warn(acpi_device_handle(device), "cannot get path name\n");
 		return;
include/acpi/acpi_bus.h +1 -0
index 9a7729e96d14c..3c4a1daed33f1 100644
@@ -361,6 +361,7 @@ struct acpi_device_data {
 struct acpi_gpio_mapping;
 
 enum acpi_device_swnode_dev_props {
+	ACPI_DEVICE_SWNODE_DEV_ROTATION,
 	ACPI_DEVICE_SWNODE_DEV_NUM_OF,
 	ACPI_DEVICE_SWNODE_DEV_NUM_ENTRIES
 };
Signature · 1 line
2.30.2
author reviewer
#20 lore raw

PATCH v3 6/8 ACPI: property: Rename parsed MIPI DisCo for Imaging properties

In reply to Sakari Ailus

MIPI DisCo for Imaging defines properties for sensor-adjacent devices such as EEPROM, LED flash or lens VCM as either device or sub-node references. This is compliant with existing DT definitions apart from property names.

Rename parsed MIPI-defined properties so drivers will have a unified view of them as defined in DT and already parsed by drivers. This can be done in-place as the MIPI-defined property strings are always longer than the DT one. This also results in loss of constness in parser function arguments.

Individual bindings to devices could define the references differently between MIPI DisCo for Imaging and DT, in terms of device or sub-node references. This will still need to be handled in the drivers themselves.

---
 drivers/acpi/internal.h |  1 +
 drivers/acpi/mipi.c     | 37 +++++++++++++++++++++++++++++++++++++
 drivers/acpi/property.c | 22 ++++++++++++----------
 3 files changed, 50 insertions(+), 10 deletions(-)
drivers/acpi/internal.h +1 -0
index fac87404e294c..f107094bfe16f 100644
@@ -289,5 +289,6 @@ static inline void acpi_init_lpit(void) { }
 void acpi_crs_csi2_swnodes_del_free(void);
 void acpi_bus_scan_crs_csi2(acpi_handle handle);
 void acpi_init_swnodes(struct acpi_device *device);
+void acpi_properties_prepare_mipi(union acpi_object *elements);
 
 #endif /* _ACPI_INTERNAL_H_ */
drivers/acpi/mipi.c +37 -0
index b81bfa1edd8ad..4211b53b41682 100644
@@ -687,3 +687,40 @@ void acpi_init_swnodes(struct acpi_device *device)
 	primary = acpi_fwnode_handle(device);
 	primary->secondary = software_node_fwnode(ads->nodes);
 }
+
+static const struct mipi_disco_prop {
+	const char *mipi_prop;
+	const char *dt_prop;
+} mipi_disco_props[] = {
+	{ "mipi-img-lens-focus",	"lens-focus" },
+	{ "mipi-img-flash-leds",	"flash-leds" },
+	{ "mipi-img-clock-frequency",	"clock-frequency" },
+	{ "mipi-img-led-max-current",	"led-max-microamp" },
+	{ "mipi-img-flash-max-current",	"flash-max-microamp" },
+	{ "mipi-img-flash-max-timeout",	"flash-max-timeout-us" },
+};
+
+/**
+ * acpi_properties_prepare_mipi - Rename MIPI properties as commin DT ones
+ *
+ * @elements: ACPI object containing _DSD properties for a device node
+ *
+ * Renames MIPI-defined properties as common DT ones. The pre-requisite is that
+ * the names of all such MIPI properties are no longer than the corresponding DT
+ * ones.
+ */
+void acpi_properties_prepare_mipi(union acpi_object *elements)
+{
+	unsigned int i;
+
+	/* Replace MIPI DisCo for Imaging property names with DT equivalents. */
+	for (i = 0; i < ARRAY_SIZE(mipi_disco_props); i++) {
+		if (!strcmp(mipi_disco_props[i].mipi_prop,
+			    elements[0].string.pointer)) {
+			WARN_ON(strscpy(elements[0].string.pointer,
+					mipi_disco_props[i].dt_prop,
+					elements[0].string.length) < 0);
+			break;
+		}
+	}
+}
drivers/acpi/property.c +12 -10
index 2b69b25c405c5..0f1dd9cc525bc 100644
@@ -124,7 +124,7 @@ static bool acpi_nondev_subnode_extract(union acpi_object *desc,
 }
 
 static bool acpi_nondev_subnode_data_ok(acpi_handle handle,
-					const union acpi_object *link,
+					union acpi_object *link,
 					struct list_head *list,
 					struct fwnode_handle *parent)
 {
@@ -145,7 +145,7 @@ static bool acpi_nondev_subnode_data_ok(acpi_handle handle,
 }
 
 static bool acpi_nondev_subnode_ok(acpi_handle scope,
-				   const union acpi_object *link,
+				   union acpi_object *link,
 				   struct list_head *list,
 				   struct fwnode_handle *parent)
 {
@@ -276,22 +276,24 @@ static bool acpi_property_value_ok(const union acpi_object *value)
 	return false;
 }
 
-static bool acpi_properties_format_valid(const union acpi_object *properties)
+static bool acpi_properties_prepare(union acpi_object *properties)
 {
-	int i;
+	unsigned int i;
 
 	for (i = 0; i < properties->package.count; i++) {
-		const union acpi_object *property;
+		union acpi_object *property = &properties->package.elements[i];
+		union acpi_object *elements = property->package.elements;
 
-		property = &properties->package.elements[i];
 		/*
 		 * Only two elements allowed, the first one must be a string and
 		 * the second one has to satisfy certain conditions.
 		 */
-		if (property->package.count != 2
-		    || property->package.elements[0].type != ACPI_TYPE_STRING
-		    || !acpi_property_value_ok(&property->package.elements[1]))
+		if (property->package.count != 2 ||
+		    elements[0].type != ACPI_TYPE_STRING ||
+		    !acpi_property_value_ok(&elements[1]))
 			return false;
+
+		acpi_properties_prepare_mipi(elements);
 	}
 	return true;
 }
@@ -523,7 +525,7 @@ static bool acpi_extract_properties(acpi_handle scope, union acpi_object *desc,
 		 * We found the matching GUID. Now validate the format of the
 		 * package immediately following it.
 		 */
-		if (!acpi_properties_format_valid(properties))
+		if (!acpi_properties_prepare(properties))
 			continue;
 
 		acpi_data_add_props(data, (const guid_t *)guid->buffer.pointer,
Signature · 1 line
2.30.2
author reviewer
#21 lore raw

PATCH v3 7/8 ACPI: property: Skip MIPI property table without "mipi-img" prefix

In reply to Sakari Ailus

For all _DSD properties, skip going through the MIPI DisCo for Imaging property name substitution table if the property doesn't have "mipi-img-" prefix.

---
 drivers/acpi/mipi.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/acpi/mipi.c +5 -0
index 4211b53b41682..17fcc0417e27d 100644
@@ -688,6 +688,8 @@ void acpi_init_swnodes(struct acpi_device *device)
 	primary->secondary = software_node_fwnode(ads->nodes);
 }
 
+#define MIPI_IMG_PREFIX "mipi-img-"
+
 static const struct mipi_disco_prop {
 	const char *mipi_prop;
 	const char *dt_prop;
@@ -713,6 +715,9 @@ void acpi_properties_prepare_mipi(union acpi_object *elements)
 {
 	unsigned int i;
 
+	if (!str_has_prefix(elements[0].string.pointer, MIPI_IMG_PREFIX))
+		return;
+
 	/* Replace MIPI DisCo for Imaging property names with DT equivalents. */
 	for (i = 0; i < ARRAY_SIZE(mipi_disco_props); i++) {
 		if (!strcmp(mipi_disco_props[i].mipi_prop,
Signature · 1 line
2.30.2
author reviewer
#22 lore raw

PATCH v3 8/8 ACPI: property: Document _CRS CSI-2 and DisCo for Imaging support

In reply to Sakari Ailus

Document how ACPI _CRS CSI-2 and DisCo for Imaging works. It's non-trivial so such documentation can be useful.

---
 drivers/acpi/mipi.c | 37 +++++++++++++++++++++++++++++++++++++
 1 file changed, 37 insertions(+)
drivers/acpi/mipi.c +37 -0
index 17fcc0417e27d..e87cfb584e39a 100644
@@ -3,6 +3,43 @@
  * MIPI DisCo for Imaging support.
  *
  * Copyright (C) 2023 Intel Corporation
+ *
+ * _CRS CSI-2 descriptors, as defined starting from ACPI 6.4 [1], contain
+ * information on cross-device CSI-2 bus configuration. The descriptors are
+ * located under transmitter devices, and the receiver devices have no direct
+ * way to access them even if the information in these descriptors is equally
+ * important for receivers. This information is amended with MIPI DisCo for
+ * Imaging [2] specification that defines _DSD data nodes and properties.
+ *
+ * The support for these is based on two-fold approach, firstly renaming
+ * properties where semantics matches and secondly gathering information to
+ * generate properties using information gathered from various sources. The
+ * former is trivial (see acpi_properties_prepare_mipi() at the end of the
+ * file) whereas the latter requires a more elaborate explanation.
+ *
+ * acpi_bus_scan_crs_csi2() scans an ACPI bus for devices with _CRS CSI-2
+ * descriptors and stores them to a linked list. This is done as traversing just
+ * this list is much smaller task than the entire DSDT. This list is then used
+ * to figure out how much memory is needed for swnodes related to a given ACPI
+ * device (handle). Further on, the same function sets the property values for
+ * the properties the values of which are obtained from the _CRS CSI-2
+ * descriptor. The information is stored into another list where the information
+ * can be looked up based on device's acpi_handle as the struct acpi_device
+ * isn't available yet at this point (and could not, as cross-device references
+ * need to be set up before the devices are available for drivers to probe).
+ *
+ * For each struct acpi_device, acpi_init_swnodes() further obtains information
+ * required to find out the values for the rest of the properties needed by
+ * drivers. This includes all port and endpoint properties as the node
+ * structures used by DT graphs and DisCo for Imaging are different. Finally the
+ * function registers software nodes for the device and sets the secondary
+ * pointer for the ACPI device's fwnode.
+ *
+ * Access to data the structures is serialised using acpi_scan_lock in scan.c.
+ *
+ * [1] https://uefi.org/sites/default/files/resources/ACPI_Spec_6_4_Jan22.pdf
+ *
+ * [2] https://www.mipi.org/specifications/mipi-disco-imaging
  */
 
 #include <linux/acpi.h>
Signature · 1 line
2.30.2