[U-Boot] [PATCH 1/3] core: Add function to get device for ofnode
It's sometimes useful to get the device associated with a given ofnode. Implement a function to implement this lookup operation.
Signed-off-by: Mario Six mario.six@gdsys.cc --- drivers/core/ofnode.c | 15 +++++++++++++++ include/dm/ofnode.h | 8 ++++++++ 2 files changed, 23 insertions(+)
diff --git a/drivers/core/ofnode.c b/drivers/core/ofnode.c index 4e4532651f..ca002063b3 100644 --- a/drivers/core/ofnode.c +++ b/drivers/core/ofnode.c @@ -16,6 +16,21 @@ #include <linux/err.h> #include <linux/ioport.h>
+struct udevice *ofnode_dev(ofnode node) +{ + struct uclass *uc; + struct udevice *dev; + + list_for_each_entry(uc, &gd->uclass_root, sibling_node) { + list_for_each_entry(dev, &uc->dev_head, uclass_node) { + if (ofnode_equal(dev_ofnode(dev), node)) + return dev; + } + } + + return NULL; +} + int ofnode_read_u32(ofnode node, const char *propname, u32 *outp) { assert(ofnode_valid(node)); diff --git a/include/dm/ofnode.h b/include/dm/ofnode.h index 0d008404f9..aec205eb80 100644 --- a/include/dm/ofnode.h +++ b/include/dm/ofnode.h @@ -193,6 +193,14 @@ static inline ofnode ofnode_null(void) return node; }
+/** + * ofnode_dev() - Get the device associated with a given ofnode + * + * @node: valid node reference to get the corresponding device for + * @return a pointer to the udevice if OK, NULL on error + */ +struct udevice *ofnode_dev(ofnode node); + /** * ofnode_read_u32() - Read a 32-bit integer from a property *
Implement a set of functions to manipulate properties in a live device tree:
* ofnode_set_property() to set generic properties of a node * ofnode_write_string() to set string properties of a node * ofnode_enable() to enable a node * ofnode_disable() to disable a node
Signed-off-by: Mario Six mario.six@gdsys.cc --- drivers/core/ofnode.c | 58 +++++++++++++++++++++++++++++++++++++++++++++++++++ include/dm/ofnode.h | 49 +++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 107 insertions(+)
diff --git a/drivers/core/ofnode.c b/drivers/core/ofnode.c index ca002063b3..90d05aa559 100644 --- a/drivers/core/ofnode.c +++ b/drivers/core/ofnode.c @@ -699,3 +699,61 @@ u64 ofnode_translate_address(ofnode node, const fdt32_t *in_addr) else return fdt_translate_address(gd->fdt_blob, ofnode_to_offset(node), in_addr); } + +#ifdef CONFIG_OF_LIVE +int ofnode_set_property(ofnode node, const char *propname, int len, + const void *value) +{ + const struct device_node *np = ofnode_to_np(node); + struct property *pp; + struct property *new; + + if (!np) + return -EINVAL; + + for (pp = np->properties; pp; pp = pp->next) { + if (strcmp(pp->name, propname) == 0) { + /* Property exists -> change value */ + pp->value = (void *)value; + pp->length = len; + return 0; + } + } + + /* Property does not exist -> append new property */ + new = malloc(sizeof(struct property)); + + new->name = strdup(propname); + new->value = malloc(len); + memcpy(new->value, value, len); + new->length = len; + new->next = NULL; + + pp->next = new; + + return 0; +} + +int ofnode_write_string(ofnode node, const char *propname, const char *value) +{ + assert(ofnode_valid(node)); + debug("%s: %s = %s", __func__, propname, value); + + return ofnode_set_property(node, propname, strlen(value) + 1, value); +} + +int ofnode_enable(ofnode node) +{ + assert(ofnode_valid(node)); + + return ofnode_write_string(node, "status", "okay"); +} + +int ofnode_disable(ofnode node) +{ + assert(ofnode_valid(node)); + + return ofnode_write_string(node, "status", "disable"); +} + +#endif diff --git a/include/dm/ofnode.h b/include/dm/ofnode.h index aec205eb80..e82ebf19c5 100644 --- a/include/dm/ofnode.h +++ b/include/dm/ofnode.h @@ -689,4 +689,53 @@ int ofnode_read_resource_byname(ofnode node, const char *name, * @return the translated address; OF_BAD_ADDR on error */ u64 ofnode_translate_address(ofnode node, const fdt32_t *in_addr); + +#ifdef CONFIG_OF_LIVE + +/** + * ofnode_set_property() - Set a property of a ofnode + * + * @node: The node for whose property should be set + * @propname: The name of the property to set + * @len: The length of the new value of the property + * @value: The new value of the property + * @return 0 if successful, -ve on error + */ +int ofnode_set_property(ofnode node, const char *propname, int len, + const void *value); + +/** + * ofnode_write_string() - Set a string property of a ofnode + * + * @node: The node for whose string property should be set + * @propname: The name of the string property to set + * @value: The new value of the string property + * @return 0 if successful, -ve on error + */ +int ofnode_write_string(ofnode node, const char *propname, const char *value); + +/** + * ofnode_enable() - Enable a device tree node given by its ofnode + * + * This function effectively sets the node's "status" property to "okay", hence + * making it available for driver model initialization. + * + * @node: The node to enable + * @return 0 if successful, -ve on error + */ +int ofnode_enable(ofnode node); + +/** + * ofnode_disable() - Disable a device tree node given by its ofnode + * + * This function effectively sets the node's "status" property to "disable", + * hence stopping it from being picked up by driver model initialization. + * + * @node: The node to disable + * @return 0 if successful, -ve on error + */ +int ofnode_disable(ofnode node); + +#endif + #endif
Hi Mario,
On 28 March 2018 at 20:37, Mario Six mario.six@gdsys.cc wrote:
Implement a set of functions to manipulate properties in a live device tree:
- ofnode_set_property() to set generic properties of a node
- ofnode_write_string() to set string properties of a node
- ofnode_enable() to enable a node
- ofnode_disable() to disable a node
Please add a test for these functions.
Signed-off-by: Mario Six mario.six@gdsys.cc
drivers/core/ofnode.c | 58 +++++++++++++++++++++++++++++++++++++++++++++++++++ include/dm/ofnode.h | 49 +++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 107 insertions(+)
diff --git a/drivers/core/ofnode.c b/drivers/core/ofnode.c index ca002063b3..90d05aa559 100644 --- a/drivers/core/ofnode.c +++ b/drivers/core/ofnode.c @@ -699,3 +699,61 @@ u64 ofnode_translate_address(ofnode node, const fdt32_t *in_addr) else return fdt_translate_address(gd->fdt_blob, ofnode_to_offset(node), in_addr); }
+#ifdef CONFIG_OF_LIVE
Is this needed?
+int ofnode_set_property(ofnode node, const char *propname, int len,
const void *value)+{
const struct device_node *np = ofnode_to_np(node);struct property *pp;struct property *new;if (!np)return -EINVAL;for (pp = np->properties; pp; pp = pp->next) {if (strcmp(pp->name, propname) == 0) {/* Property exists -> change value */pp->value = (void *)value;pp->length = len;return 0;}}/* Property does not exist -> append new property */new = malloc(sizeof(struct property));new->name = strdup(propname);new->value = malloc(len);memcpy(new->value, value, len);new->length = len;new->next = NULL;pp->next = new;return 0;+}
+int ofnode_write_string(ofnode node, const char *propname, const char *value) +{
assert(ofnode_valid(node));
What does this function do if livetree is not enabled?
debug("%s: %s = %s", __func__, propname, value);return ofnode_set_property(node, propname, strlen(value) + 1, value);+}
+int ofnode_enable(ofnode node) +{
assert(ofnode_valid(node));return ofnode_write_string(node, "status", "okay");+}
+int ofnode_disable(ofnode node) +{
assert(ofnode_valid(node));return ofnode_write_string(node, "status", "disable");+}
+#endif diff --git a/include/dm/ofnode.h b/include/dm/ofnode.h index aec205eb80..e82ebf19c5 100644 --- a/include/dm/ofnode.h +++ b/include/dm/ofnode.h @@ -689,4 +689,53 @@ int ofnode_read_resource_byname(ofnode node, const char *name,
- @return the translated address; OF_BAD_ADDR on error
*/ u64 ofnode_translate_address(ofnode node, const fdt32_t *in_addr);
+#ifdef CONFIG_OF_LIVE
+/**
- ofnode_set_property() - Set a property of a ofnode
- @node: The node for whose property should be set
- @propname: The name of the property to set
- @len: The length of the new value of the property
- @value: The new value of the property
- @return 0 if successful, -ve on error
- */
+int ofnode_set_property(ofnode node, const char *propname, int len,
const void *value);
We should think about consistency here. Maybe
ofnode_write_prop()?
+/**
- ofnode_write_string() - Set a string property of a ofnode
- @node: The node for whose string property should be set
- @propname: The name of the string property to set
- @value: The new value of the string property
- @return 0 if successful, -ve on error
- */
+int ofnode_write_string(ofnode node, const char *propname, const char *value);
+/**
- ofnode_enable() - Enable a device tree node given by its ofnode
- This function effectively sets the node's "status" property to "okay", hence
- making it available for driver model initialization.
- @node: The node to enable
- @return 0 if successful, -ve on error
- */
+int ofnode_enable(ofnode node);
+/**
- ofnode_disable() - Disable a device tree node given by its ofnode
- This function effectively sets the node's "status" property to "disable",
- hence stopping it from being picked up by driver model initialization.
- @node: The node to disable
- @return 0 if successful, -ve on error
- */
+int ofnode_disable(ofnode node);
Would it be OK to have one function like ofnode_set_enabled(ofnode node, bool enable) ?
Regards, Simon
Hi Simon,
On Fri, Mar 30, 2018 at 12:43 AM, Simon Glass sjg@chromium.org wrote:
Hi Mario,
On 28 March 2018 at 20:37, Mario Six mario.six@gdsys.cc wrote:
Implement a set of functions to manipulate properties in a live device tree:
- ofnode_set_property() to set generic properties of a node
- ofnode_write_string() to set string properties of a node
- ofnode_enable() to enable a node
- ofnode_disable() to disable a node
Please add a test for these functions.
Signed-off-by: Mario Six mario.six@gdsys.cc
drivers/core/ofnode.c | 58 +++++++++++++++++++++++++++++++++++++++++++++++++++ include/dm/ofnode.h | 49 +++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 107 insertions(+)
diff --git a/drivers/core/ofnode.c b/drivers/core/ofnode.c index ca002063b3..90d05aa559 100644 --- a/drivers/core/ofnode.c +++ b/drivers/core/ofnode.c @@ -699,3 +699,61 @@ u64 ofnode_translate_address(ofnode node, const fdt32_t *in_addr) else return fdt_translate_address(gd->fdt_blob, ofnode_to_offset(node), in_addr); }
+#ifdef CONFIG_OF_LIVE
Is this needed?
See below, these functions don't make sense if there is no livetree.
+int ofnode_set_property(ofnode node, const char *propname, int len,
const void *value)+{
const struct device_node *np = ofnode_to_np(node);struct property *pp;struct property *new;if (!np)return -EINVAL;for (pp = np->properties; pp; pp = pp->next) {if (strcmp(pp->name, propname) == 0) {/* Property exists -> change value */pp->value = (void *)value;pp->length = len;return 0;}}/* Property does not exist -> append new property */new = malloc(sizeof(struct property));new->name = strdup(propname);new->value = malloc(len);memcpy(new->value, value, len);new->length = len;new->next = NULL;pp->next = new;return 0;+}
+int ofnode_write_string(ofnode node, const char *propname, const char *value) +{
assert(ofnode_valid(node));What does this function do if livetree is not enabled?
The idea was to not make them available if livetree is not enabled (hence the #ifdef CONFIG_OF_LIVE).
Making them nops in case livetree is not available would work as well if that's preferable.
debug("%s: %s = %s", __func__, propname, value);return ofnode_set_property(node, propname, strlen(value) + 1, value);+}
+int ofnode_enable(ofnode node) +{
assert(ofnode_valid(node));return ofnode_write_string(node, "status", "okay");+}
+int ofnode_disable(ofnode node) +{
assert(ofnode_valid(node));return ofnode_write_string(node, "status", "disable");+}
+#endif diff --git a/include/dm/ofnode.h b/include/dm/ofnode.h index aec205eb80..e82ebf19c5 100644 --- a/include/dm/ofnode.h +++ b/include/dm/ofnode.h @@ -689,4 +689,53 @@ int ofnode_read_resource_byname(ofnode node, const char *name,
- @return the translated address; OF_BAD_ADDR on error
*/ u64 ofnode_translate_address(ofnode node, const fdt32_t *in_addr);
+#ifdef CONFIG_OF_LIVE
+/**
- ofnode_set_property() - Set a property of a ofnode
- @node: The node for whose property should be set
- @propname: The name of the property to set
- @len: The length of the new value of the property
- @value: The new value of the property
- @return 0 if successful, -ve on error
- */
+int ofnode_set_property(ofnode node, const char *propname, int len,
const void *value);We should think about consistency here. Maybe
ofnode_write_prop()?
+/**
- ofnode_write_string() - Set a string property of a ofnode
- @node: The node for whose string property should be set
- @propname: The name of the string property to set
- @value: The new value of the string property
- @return 0 if successful, -ve on error
- */
+int ofnode_write_string(ofnode node, const char *propname, const char *value);
+/**
- ofnode_enable() - Enable a device tree node given by its ofnode
- This function effectively sets the node's "status" property to "okay", hence
- making it available for driver model initialization.
- @node: The node to enable
- @return 0 if successful, -ve on error
- */
+int ofnode_enable(ofnode node);
+/**
- ofnode_disable() - Disable a device tree node given by its ofnode
- This function effectively sets the node's "status" property to "disable",
- hence stopping it from being picked up by driver model initialization.
- @node: The node to disable
- @return 0 if successful, -ve on error
- */
+int ofnode_disable(ofnode node);
Would it be OK to have one function like ofnode_set_enabled(ofnode node, bool enable) ?
Regards, Simon
Everything else will be addressed in v2. Thanks for reviewing!
Best regards,
Mario
Hi Mario,
On 10 April 2018 at 05:23, Mario Six mario.six@gdsys.cc wrote:
Hi Simon,
On Fri, Mar 30, 2018 at 12:43 AM, Simon Glass sjg@chromium.org wrote:
Hi Mario,
On 28 March 2018 at 20:37, Mario Six mario.six@gdsys.cc wrote:
Implement a set of functions to manipulate properties in a live device tree:
- ofnode_set_property() to set generic properties of a node
- ofnode_write_string() to set string properties of a node
- ofnode_enable() to enable a node
- ofnode_disable() to disable a node
Please add a test for these functions.
Signed-off-by: Mario Six mario.six@gdsys.cc
drivers/core/ofnode.c | 58 +++++++++++++++++++++++++++++++++++++++++++++++++++ include/dm/ofnode.h | 49 +++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 107 insertions(+)
diff --git a/drivers/core/ofnode.c b/drivers/core/ofnode.c index ca002063b3..90d05aa559 100644 --- a/drivers/core/ofnode.c +++ b/drivers/core/ofnode.c @@ -699,3 +699,61 @@ u64 ofnode_translate_address(ofnode node, const fdt32_t *in_addr) else return fdt_translate_address(gd->fdt_blob, ofnode_to_offset(node), in_addr); }
+#ifdef CONFIG_OF_LIVE
Is this needed?
See below, these functions don't make sense if there is no livetree.
Right, but they should still compile?
+int ofnode_set_property(ofnode node, const char *propname, int len,
const void *value)+{
const struct device_node *np = ofnode_to_np(node);struct property *pp;struct property *new;if (!np)return -EINVAL;for (pp = np->properties; pp; pp = pp->next) {if (strcmp(pp->name, propname) == 0) {/* Property exists -> change value */pp->value = (void *)value;pp->length = len;return 0;}}/* Property does not exist -> append new property */new = malloc(sizeof(struct property));new->name = strdup(propname);new->value = malloc(len);memcpy(new->value, value, len);new->length = len;new->next = NULL;pp->next = new;return 0;+}
+int ofnode_write_string(ofnode node, const char *propname, const char *value) +{
assert(ofnode_valid(node));What does this function do if livetree is not enabled?
The idea was to not make them available if livetree is not enabled (hence the #ifdef CONFIG_OF_LIVE).
Making them nops in case livetree is not available would work as well if that's preferable.
Yes. Then they could return -ENOSYS, for example. Then we at least have a consistent API for both live/flat tree, instead of them splitting away from each other.
debug("%s: %s = %s", __func__, propname, value);return ofnode_set_property(node, propname, strlen(value) + 1, value);+}
+int ofnode_enable(ofnode node) +{
assert(ofnode_valid(node));return ofnode_write_string(node, "status", "okay");+}
+int ofnode_disable(ofnode node) +{
assert(ofnode_valid(node));return ofnode_write_string(node, "status", "disable");+}
+#endif diff --git a/include/dm/ofnode.h b/include/dm/ofnode.h index aec205eb80..e82ebf19c5 100644 --- a/include/dm/ofnode.h +++ b/include/dm/ofnode.h @@ -689,4 +689,53 @@ int ofnode_read_resource_byname(ofnode node, const char *name,
- @return the translated address; OF_BAD_ADDR on error
*/ u64 ofnode_translate_address(ofnode node, const fdt32_t *in_addr);
+#ifdef CONFIG_OF_LIVE
+/**
- ofnode_set_property() - Set a property of a ofnode
- @node: The node for whose property should be set
- @propname: The name of the property to set
- @len: The length of the new value of the property
- @value: The new value of the property
- @return 0 if successful, -ve on error
- */
+int ofnode_set_property(ofnode node, const char *propname, int len,
const void *value);We should think about consistency here. Maybe
ofnode_write_prop()?
Yes
+/**
- ofnode_write_string() - Set a string property of a ofnode
- @node: The node for whose string property should be set
- @propname: The name of the string property to set
- @value: The new value of the string property
- @return 0 if successful, -ve on error
- */
+int ofnode_write_string(ofnode node, const char *propname, const char *value);
+/**
- ofnode_enable() - Enable a device tree node given by its ofnode
- This function effectively sets the node's "status" property to "okay", hence
- making it available for driver model initialization.
- @node: The node to enable
- @return 0 if successful, -ve on error
- */
+int ofnode_enable(ofnode node);
+/**
- ofnode_disable() - Disable a device tree node given by its ofnode
- This function effectively sets the node's "status" property to "disable",
- hence stopping it from being picked up by driver model initialization.
- @node: The node to disable
- @return 0 if successful, -ve on error
- */
+int ofnode_disable(ofnode node);
Would it be OK to have one function like ofnode_set_enabled(ofnode node, bool enable) ?
Regards, Simon
Everything else will be addressed in v2. Thanks for reviewing!
Best regards,
Mario
Regards, Simon
Hi Simon,
On Thu, Apr 12, 2018 at 6:42 PM, Simon Glass sjg@chromium.org wrote:
Hi Mario,
On 10 April 2018 at 05:23, Mario Six mario.six@gdsys.cc wrote:
Hi Simon,
On Fri, Mar 30, 2018 at 12:43 AM, Simon Glass sjg@chromium.org wrote:
Hi Mario,
On 28 March 2018 at 20:37, Mario Six mario.six@gdsys.cc wrote:
Implement a set of functions to manipulate properties in a live device tree:
- ofnode_set_property() to set generic properties of a node
- ofnode_write_string() to set string properties of a node
- ofnode_enable() to enable a node
- ofnode_disable() to disable a node
Please add a test for these functions.
Signed-off-by: Mario Six mario.six@gdsys.cc
drivers/core/ofnode.c | 58 +++++++++++++++++++++++++++++++++++++++++++++++++++ include/dm/ofnode.h | 49 +++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 107 insertions(+)
diff --git a/drivers/core/ofnode.c b/drivers/core/ofnode.c index ca002063b3..90d05aa559 100644 --- a/drivers/core/ofnode.c +++ b/drivers/core/ofnode.c @@ -699,3 +699,61 @@ u64 ofnode_translate_address(ofnode node, const fdt32_t *in_addr) else return fdt_translate_address(gd->fdt_blob, ofnode_to_offset(node), in_addr); }
+#ifdef CONFIG_OF_LIVE
Is this needed?
See below, these functions don't make sense if there is no livetree.
Right, but they should still compile?
+int ofnode_set_property(ofnode node, const char *propname, int len,
const void *value)+{
const struct device_node *np = ofnode_to_np(node);struct property *pp;struct property *new;if (!np)return -EINVAL;for (pp = np->properties; pp; pp = pp->next) {if (strcmp(pp->name, propname) == 0) {/* Property exists -> change value */pp->value = (void *)value;pp->length = len;return 0;}}/* Property does not exist -> append new property */new = malloc(sizeof(struct property));new->name = strdup(propname);new->value = malloc(len);memcpy(new->value, value, len);new->length = len;new->next = NULL;pp->next = new;return 0;+}
+int ofnode_write_string(ofnode node, const char *propname, const char *value) +{
assert(ofnode_valid(node));What does this function do if livetree is not enabled?
The idea was to not make them available if livetree is not enabled (hence the #ifdef CONFIG_OF_LIVE).
Making them nops in case livetree is not available would work as well if that's preferable.
Yes. Then they could return -ENOSYS, for example. Then we at least have a consistent API for both live/flat tree, instead of them splitting away from each other.
OK, I'll use that approach in v2.
debug("%s: %s = %s", __func__, propname, value);return ofnode_set_property(node, propname, strlen(value) + 1, value);+}
+int ofnode_enable(ofnode node) +{
assert(ofnode_valid(node));return ofnode_write_string(node, "status", "okay");+}
+int ofnode_disable(ofnode node) +{
assert(ofnode_valid(node));return ofnode_write_string(node, "status", "disable");+}
+#endif diff --git a/include/dm/ofnode.h b/include/dm/ofnode.h index aec205eb80..e82ebf19c5 100644 --- a/include/dm/ofnode.h +++ b/include/dm/ofnode.h @@ -689,4 +689,53 @@ int ofnode_read_resource_byname(ofnode node, const char *name,
- @return the translated address; OF_BAD_ADDR on error
*/ u64 ofnode_translate_address(ofnode node, const fdt32_t *in_addr);
+#ifdef CONFIG_OF_LIVE
+/**
- ofnode_set_property() - Set a property of a ofnode
- @node: The node for whose property should be set
- @propname: The name of the property to set
- @len: The length of the new value of the property
- @value: The new value of the property
- @return 0 if successful, -ve on error
- */
+int ofnode_set_property(ofnode node, const char *propname, int len,
const void *value);We should think about consistency here. Maybe
ofnode_write_prop()?
Yes
+/**
- ofnode_write_string() - Set a string property of a ofnode
- @node: The node for whose string property should be set
- @propname: The name of the string property to set
- @value: The new value of the string property
- @return 0 if successful, -ve on error
- */
+int ofnode_write_string(ofnode node, const char *propname, const char *value);
+/**
- ofnode_enable() - Enable a device tree node given by its ofnode
- This function effectively sets the node's "status" property to "okay", hence
- making it available for driver model initialization.
- @node: The node to enable
- @return 0 if successful, -ve on error
- */
+int ofnode_enable(ofnode node);
+/**
- ofnode_disable() - Disable a device tree node given by its ofnode
- This function effectively sets the node's "status" property to "disable",
- hence stopping it from being picked up by driver model initialization.
- @node: The node to disable
- @return 0 if successful, -ve on error
- */
+int ofnode_disable(ofnode node);
Would it be OK to have one function like ofnode_set_enabled(ofnode node, bool enable) ?
Regards, Simon
Everything else will be addressed in v2. Thanks for reviewing!
Best regards,
Mario
Regards, Simon
Best regards, Mario
We cannot use device structures to disable devices, since getting them with the API functions would bind and activate the device, which would fail if the underlying device does not exist.
Hence, add a function to disable devices by path in a live device tree.
Signed-off-by: Mario Six mario.six@gdsys.cc --- drivers/core/device.c | 36 ++++++++++++++++++++++++++++++++++++ include/dm/device.h | 20 ++++++++++++++++++++ 2 files changed, 56 insertions(+)
diff --git a/drivers/core/device.c b/drivers/core/device.c index 940a153c58..c627453bb9 100644 --- a/drivers/core/device.c +++ b/drivers/core/device.c @@ -724,3 +724,39 @@ bool of_machine_is_compatible(const char *compat)
return !fdt_node_check_compatible(fdt, 0, compat); } + +#ifdef CONFIG_OF_LIVE +int dev_disable_by_path(const char *path) +{ + ofnode node = np_to_ofnode(of_find_node_by_path(path)); + struct udevice *dev = ofnode_dev(node); + int res; + + res = device_remove(dev, DM_REMOVE_NORMAL); + if (res) + return res; + + res = device_unbind(dev); + if (res) + return res; + + return ofnode_disable(node); +} + +int dev_enable_by_path(const char *path) +{ + ofnode node = np_to_ofnode(of_find_node_by_path(path)); + ofnode pnode = ofnode_get_parent(node); + struct udevice *parent = ofnode_dev(pnode); + int res; + + if (!parent) + return -EINVAL; + + res = ofnode_enable(node); + if (res) + return res; + + return lists_bind_fdt(parent, node, NULL); +} +#endif /* CONFIG_OF_LIVE */ diff --git a/include/dm/device.h b/include/dm/device.h index 7786b1cf4e..f55907966a 100644 --- a/include/dm/device.h +++ b/include/dm/device.h @@ -586,6 +586,26 @@ bool device_is_compatible(struct udevice *dev, const char *compat); */ bool of_machine_is_compatible(const char *compat);
+#ifdef CONFIG_OF_LIVE + +/** + * dev_disable_by_path() - Disable a device given its device tree path + * + * @path: The device tree path identifying the device to be disabled + * @return 0 on success, -ve on error + */ +int dev_disable_by_path(const char *path); + +/** + * dev_enable_by_path() - Enable a device given its device tree path + * + * @path: The device tree path identifying the device to be enabled + * @return 0 on success, -ve on error + */ +int dev_enable_by_path(const char *path); + +#endif /* CONFIG_OF_LIVE */ + /** * device_is_on_pci_bus - Test if a device is on a PCI bus *
Hi Mario,
On 28 March 2018 at 20:37, Mario Six mario.six@gdsys.cc wrote:
We cannot use device structures to disable devices, since getting them with the API functions would bind and activate the device, which would fail if the underlying device does not exist.
Hence, add a function to disable devices by path in a live device tree.
What is the use case here? Is it to disable something after it has already been bound / probed? Why can this not be done in the device tree before U-Boot starts?
Also this needs a test.
Signed-off-by: Mario Six mario.six@gdsys.cc
drivers/core/device.c | 36 ++++++++++++++++++++++++++++++++++++ include/dm/device.h | 20 ++++++++++++++++++++ 2 files changed, 56 insertions(+)
diff --git a/drivers/core/device.c b/drivers/core/device.c index 940a153c58..c627453bb9 100644 --- a/drivers/core/device.c +++ b/drivers/core/device.c @@ -724,3 +724,39 @@ bool of_machine_is_compatible(const char *compat)
return !fdt_node_check_compatible(fdt, 0, compat);}
+#ifdef CONFIG_OF_LIVE +int dev_disable_by_path(const char *path) +{
ofnode node = np_to_ofnode(of_find_node_by_path(path));
Please see ofnode_path()
struct udevice *dev = ofnode_dev(node);int res;res = device_remove(dev, DM_REMOVE_NORMAL);if (res)return res;res = device_unbind(dev);if (res)return res;return ofnode_disable(node);+}
+int dev_enable_by_path(const char *path) +{
ofnode node = np_to_ofnode(of_find_node_by_path(path));ofnode pnode = ofnode_get_parent(node);struct udevice *parent = ofnode_dev(pnode);int res;if (!parent)return -EINVAL;res = ofnode_enable(node);if (res)return res;return lists_bind_fdt(parent, node, NULL);+} +#endif /* CONFIG_OF_LIVE */ diff --git a/include/dm/device.h b/include/dm/device.h index 7786b1cf4e..f55907966a 100644 --- a/include/dm/device.h +++ b/include/dm/device.h @@ -586,6 +586,26 @@ bool device_is_compatible(struct udevice *dev, const char *compat); */ bool of_machine_is_compatible(const char *compat);
+#ifdef CONFIG_OF_LIVE
+/**
- dev_disable_by_path() - Disable a device given its device tree path
- @path: The device tree path identifying the device to be disabled
- @return 0 on success, -ve on error
- */
+int dev_disable_by_path(const char *path);
+/**
- dev_enable_by_path() - Enable a device given its device tree path
- @path: The device tree path identifying the device to be enabled
- @return 0 on success, -ve on error
- */
+int dev_enable_by_path(const char *path);
+#endif /* CONFIG_OF_LIVE */
/**
- device_is_on_pci_bus - Test if a device is on a PCI bus
-- 2.16.1
Regards, Simon
Hi Simon,
On Fri, Mar 30, 2018 at 12:43 AM, Simon Glass sjg@chromium.org wrote:
Hi Mario,
On 28 March 2018 at 20:37, Mario Six mario.six@gdsys.cc wrote:
We cannot use device structures to disable devices, since getting them with the API functions would bind and activate the device, which would fail if the underlying device does not exist.
Hence, add a function to disable devices by path in a live device tree.
What is the use case here? Is it to disable something after it has already been bound / probed? Why can this not be done in the device tree before U-Boot starts?
The devices that may not be in the tree are all disabled in the device tree. Then, we detect which devices are actually there, and enable those that are. That way we can use a single device tree for lots of actual boards, which are all very minor variants of each other.
Also this needs a test.
Signed-off-by: Mario Six mario.six@gdsys.cc
drivers/core/device.c | 36 ++++++++++++++++++++++++++++++++++++ include/dm/device.h | 20 ++++++++++++++++++++ 2 files changed, 56 insertions(+)
diff --git a/drivers/core/device.c b/drivers/core/device.c index 940a153c58..c627453bb9 100644 --- a/drivers/core/device.c +++ b/drivers/core/device.c @@ -724,3 +724,39 @@ bool of_machine_is_compatible(const char *compat)
return !fdt_node_check_compatible(fdt, 0, compat);}
+#ifdef CONFIG_OF_LIVE +int dev_disable_by_path(const char *path) +{
ofnode node = np_to_ofnode(of_find_node_by_path(path));Please see ofnode_path()
struct udevice *dev = ofnode_dev(node);int res;res = device_remove(dev, DM_REMOVE_NORMAL);if (res)return res;res = device_unbind(dev);if (res)return res;return ofnode_disable(node);+}
+int dev_enable_by_path(const char *path) +{
ofnode node = np_to_ofnode(of_find_node_by_path(path));ofnode pnode = ofnode_get_parent(node);struct udevice *parent = ofnode_dev(pnode);int res;if (!parent)return -EINVAL;res = ofnode_enable(node);if (res)return res;return lists_bind_fdt(parent, node, NULL);+} +#endif /* CONFIG_OF_LIVE */ diff --git a/include/dm/device.h b/include/dm/device.h index 7786b1cf4e..f55907966a 100644 --- a/include/dm/device.h +++ b/include/dm/device.h @@ -586,6 +586,26 @@ bool device_is_compatible(struct udevice *dev, const char *compat); */ bool of_machine_is_compatible(const char *compat);
+#ifdef CONFIG_OF_LIVE
+/**
- dev_disable_by_path() - Disable a device given its device tree path
- @path: The device tree path identifying the device to be disabled
- @return 0 on success, -ve on error
- */
+int dev_disable_by_path(const char *path);
+/**
- dev_enable_by_path() - Enable a device given its device tree path
- @path: The device tree path identifying the device to be enabled
- @return 0 on success, -ve on error
- */
+int dev_enable_by_path(const char *path);
+#endif /* CONFIG_OF_LIVE */
/**
- device_is_on_pci_bus - Test if a device is on a PCI bus
-- 2.16.1
Regards, Simon
Everything else will be addressed in v2. Thanks for reviewing!
Best regards,
Mario
Hi Mario,
On 28 March 2018 at 20:37, Mario Six mario.six@gdsys.cc wrote:
It's sometimes useful to get the device associated with a given ofnode. Implement a function to implement this lookup operation.
Where would you use this? Can you not use phandles to find the device? Or uclass_get_device_by_ofnode() ?
Signed-off-by: Mario Six mario.six@gdsys.cc
drivers/core/ofnode.c | 15 +++++++++++++++ include/dm/ofnode.h | 8 ++++++++ 2 files changed, 23 insertions(+)
diff --git a/drivers/core/ofnode.c b/drivers/core/ofnode.c index 4e4532651f..ca002063b3 100644 --- a/drivers/core/ofnode.c +++ b/drivers/core/ofnode.c @@ -16,6 +16,21 @@ #include <linux/err.h> #include <linux/ioport.h>
+struct udevice *ofnode_dev(ofnode node)
Can you please add a test for this?
This seems like an internal function since it does not probe the device. So how about putting it in device.h:
device_get_by_ofnode() - does probe the device it returns device_find_by_ofnode() - doesn't probe
+{
struct uclass *uc;struct udevice *dev;list_for_each_entry(uc, &gd->uclass_root, sibling_node) {list_for_each_entry(dev, &uc->dev_head, uclass_node) {if (ofnode_equal(dev_ofnode(dev), node))return dev;}}return NULL;+}
int ofnode_read_u32(ofnode node, const char *propname, u32 *outp) { assert(ofnode_valid(node)); diff --git a/include/dm/ofnode.h b/include/dm/ofnode.h index 0d008404f9..aec205eb80 100644 --- a/include/dm/ofnode.h +++ b/include/dm/ofnode.h @@ -193,6 +193,14 @@ static inline ofnode ofnode_null(void) return node; }
+/**
- ofnode_dev() - Get the device associated with a given ofnode
- @node: valid node reference to get the corresponding device for
- @return a pointer to the udevice if OK, NULL on error
- */
+struct udevice *ofnode_dev(ofnode node);
/**
- ofnode_read_u32() - Read a 32-bit integer from a property
-- 2.16.1
Regards, Simon
Hi Simon,
On Fri, Mar 30, 2018 at 12:43 AM, Simon Glass sjg@chromium.org wrote:
Hi Mario,
On 28 March 2018 at 20:37, Mario Six mario.six@gdsys.cc wrote:
It's sometimes useful to get the device associated with a given ofnode. Implement a function to implement this lookup operation.
Where would you use this? Can you not use phandles to find the device? Or uclass_get_device_by_ofnode() ?
The function is used with the dev_{enable,disable}_by_path in the next patch: If I used any of the uclass_* functions or similar, the device would be probed, which is not what I want, since the device may not actually be physically present.
Signed-off-by: Mario Six mario.six@gdsys.cc
drivers/core/ofnode.c | 15 +++++++++++++++ include/dm/ofnode.h | 8 ++++++++ 2 files changed, 23 insertions(+)
diff --git a/drivers/core/ofnode.c b/drivers/core/ofnode.c index 4e4532651f..ca002063b3 100644 --- a/drivers/core/ofnode.c +++ b/drivers/core/ofnode.c @@ -16,6 +16,21 @@ #include <linux/err.h> #include <linux/ioport.h>
+struct udevice *ofnode_dev(ofnode node)
Can you please add a test for this?
This seems like an internal function since it does not probe the device. So how about putting it in device.h:
device_get_by_ofnode() - does probe the device it returns device_find_by_ofnode() - doesn't probe
+{
struct uclass *uc;struct udevice *dev;list_for_each_entry(uc, &gd->uclass_root, sibling_node) {list_for_each_entry(dev, &uc->dev_head, uclass_node) {if (ofnode_equal(dev_ofnode(dev), node))return dev;}}return NULL;+}
int ofnode_read_u32(ofnode node, const char *propname, u32 *outp) { assert(ofnode_valid(node)); diff --git a/include/dm/ofnode.h b/include/dm/ofnode.h index 0d008404f9..aec205eb80 100644 --- a/include/dm/ofnode.h +++ b/include/dm/ofnode.h @@ -193,6 +193,14 @@ static inline ofnode ofnode_null(void) return node; }
+/**
- ofnode_dev() - Get the device associated with a given ofnode
- @node: valid node reference to get the corresponding device for
- @return a pointer to the udevice if OK, NULL on error
- */
+struct udevice *ofnode_dev(ofnode node);
/**
- ofnode_read_u32() - Read a 32-bit integer from a property
-- 2.16.1
Regards, Simon
Everything else will be addressed in v2. Thanks for reviewing!
Best regards,
Mario
Hi Mario,
On 10 April 2018 at 05:34, Mario Six mario.six@gdsys.cc wrote:
Hi Simon,
On Fri, Mar 30, 2018 at 12:43 AM, Simon Glass sjg@chromium.org wrote:
Hi Mario,
On 28 March 2018 at 20:37, Mario Six mario.six@gdsys.cc wrote:
It's sometimes useful to get the device associated with a given ofnode. Implement a function to implement this lookup operation.
Where would you use this? Can you not use phandles to find the device? Or uclass_get_device_by_ofnode() ?
The function is used with the dev_{enable,disable}_by_path in the next patch: If I used any of the uclass_* functions or similar, the device would be probed, which is not what I want, since the device may not actually be physically present.
So how about using uclass_find_device_by_ofnode() ?
Regards, Simon
Hi Simon,
On Thu, Apr 12, 2018 at 6:42 PM, Simon Glass sjg@chromium.org wrote:
Hi Mario,
On 10 April 2018 at 05:34, Mario Six mario.six@gdsys.cc wrote:
Hi Simon,
On Fri, Mar 30, 2018 at 12:43 AM, Simon Glass sjg@chromium.org wrote:
Hi Mario,
On 28 March 2018 at 20:37, Mario Six mario.six@gdsys.cc wrote:
It's sometimes useful to get the device associated with a given ofnode. Implement a function to implement this lookup operation.
Where would you use this? Can you not use phandles to find the device? Or uclass_get_device_by_ofnode() ?
The function is used with the dev_{enable,disable}_by_path in the next patch: If I used any of the uclass_* functions or similar, the device would be probed, which is not what I want, since the device may not actually be physically present.
So how about using uclass_find_device_by_ofnode() ?
That would work for the disabling, true, but not for the enabling (which is what is used in the upcoming board): Since the node is declared as disabled in the DT, the device is not even bound (so uclass_find_device_by_ofnode) won't return it.
A more elegant solution would be to have device_probe check again if the underlying ofnode is disabled, and stop the probing if that's the case. In this scenario the disabled devices would still be displayed in the tree, but never probed, which is probably OK (I don't know if there would be any side effects with iterating through devices, for example). But changing the behavior of such elementary API functions is probably not a good idea.
Regards, Simon
Best regards, Mario
Hi Mario,
On 18 April 2018 at 02:20, Mario Six mario.six@gdsys.cc wrote:
Hi Simon,
On Thu, Apr 12, 2018 at 6:42 PM, Simon Glass sjg@chromium.org wrote:
Hi Mario,
On 10 April 2018 at 05:34, Mario Six mario.six@gdsys.cc wrote:
Hi Simon,
On Fri, Mar 30, 2018 at 12:43 AM, Simon Glass sjg@chromium.org wrote:
Hi Mario,
On 28 March 2018 at 20:37, Mario Six mario.six@gdsys.cc wrote:
It's sometimes useful to get the device associated with a given
ofnode.
Implement a function to implement this lookup operation.
Where would you use this? Can you not use phandles to find the device? Or uclass_get_device_by_ofnode() ?
The function is used with the dev_{enable,disable}_by_path in the next
patch:
If I used any of the uclass_* functions or similar, the device would be
probed,
which is not what I want, since the device may not actually be
physically
present.
So how about using uclass_find_device_by_ofnode() ?
That would work for the disabling, true, but not for the enabling (which
is
what is used in the upcoming board): Since the node is declared as
disabled in
the DT, the device is not even bound (so uclass_find_device_by_ofnode)
won't
return it.
A more elegant solution would be to have device_probe check again if the underlying ofnode is disabled, and stop the probing if that's the case.
In this
scenario the disabled devices would still be displayed in the tree, but
never
probed, which is probably OK (I don't know if there would be any side
effects
with iterating through devices, for example). But changing the behavior
of such
elementary API functions is probably not a good idea.
That seems to be a different topic.
Fundamentally I don't see the difference between uclass_find_device_by_ofnode() and your ofnode_dev().
If you want to enable something after probing you will have to call device_bind() or similar. If that is your intent, I think you need a different function from ofnode_dev(), since it also relies on the device already being bound.
Regards, Simon
Hi Simon,
On Sun, Apr 22, 2018 at 10:13 PM, Simon Glass sjg@chromium.org wrote:
Hi Mario,
On 18 April 2018 at 02:20, Mario Six mario.six@gdsys.cc wrote:
Hi Simon,
On Thu, Apr 12, 2018 at 6:42 PM, Simon Glass sjg@chromium.org wrote:
Hi Mario,
On 10 April 2018 at 05:34, Mario Six mario.six@gdsys.cc wrote:
Hi Simon,
On Fri, Mar 30, 2018 at 12:43 AM, Simon Glass sjg@chromium.org wrote:
Hi Mario,
On 28 March 2018 at 20:37, Mario Six mario.six@gdsys.cc wrote:
It's sometimes useful to get the device associated with a given
ofnode.
Implement a function to implement this lookup operation.
Where would you use this? Can you not use phandles to find the device? Or uclass_get_device_by_ofnode() ?
The function is used with the dev_{enable,disable}_by_path in the next
patch:
If I used any of the uclass_* functions or similar, the device would be
probed,
which is not what I want, since the device may not actually be
physically
present.
So how about using uclass_find_device_by_ofnode() ?
That would work for the disabling, true, but not for the enabling (which
is
what is used in the upcoming board): Since the node is declared as
disabled in
the DT, the device is not even bound (so uclass_find_device_by_ofnode)
won't
return it.
A more elegant solution would be to have device_probe check again if the underlying ofnode is disabled, and stop the probing if that's the case.
In this
scenario the disabled devices would still be displayed in the tree, but
never
probed, which is probably OK (I don't know if there would be any side
effects
with iterating through devices, for example). But changing the behavior
of such
elementary API functions is probably not a good idea.
That seems to be a different topic.
Fundamentally I don't see the difference between uclass_find_device_by_ofnode() and your ofnode_dev().
If you want to enable something after probing you will have to call device_bind() or similar. If that is your intent, I think you need a different function from ofnode_dev(), since it also relies on the device already being bound.
Ah, yes, I forgot that the *find* functions don't probe, sorry about that.
Yes, right, it won't be a problem to use uclass_find_device_by_ofnode instead then.
Regards, Simon
Best regards, Mario
participants (2)
-
Mario Six -
Simon Glass