| Message ID | 20231101072729.1142578-1-sakari.ailus@linux.intel.com (mailing list archive) |
|---|---|
| State | Superseded, archived |
| Headers | show |
| Series | [v2,1/1] device property: Add fwnode_name_eq() | expand |
On Wed, Nov 01, 2023 at 09:27:29AM +0200, Sakari Ailus wrote: > Add fwnode_name_eq() to implement the functionality of of_node_name_eq() > on fwnode property API. The same convention of ending the comparison at > '@' (besides '\0') is applied on also both ACPI and swnode. The function > is intended for comparing unit address-less node names on DT and firmware > or swnodes compliant with DT bindings. ... > - Use size_t type for len. Strictly speaking it should be ptrdiff_t according to the code. ... > + * the comparison to either '\0' or '@' character NUL is a human-readable substitution to less parsable '\0'.
On Wed, Nov 01, 2023 at 11:50:00AM +0200, Andy Shevchenko wrote: > On Wed, Nov 01, 2023 at 09:27:29AM +0200, Sakari Ailus wrote: > > Add fwnode_name_eq() to implement the functionality of of_node_name_eq() > > on fwnode property API. The same convention of ending the comparison at > > '@' (besides '\0') is applied on also both ACPI and swnode. The function > > is intended for comparing unit address-less node names on DT and firmware > > or swnodes compliant with DT bindings. > > ... > > > - Use size_t type for len. > > Strictly speaking it should be ptrdiff_t according to the code. It'll be compared to size_t right after. ptrdiff_t is signed and we know we have a positive number here so size_t seems like a better choice. > > ... > > > + * the comparison to either '\0' or '@' character > > NUL is a human-readable substitution to less parsable '\0'. I can send v3 with that but I'll wait a bit if there are more comments.
Hi Sakari, Thank you for the patch. On Wed, Nov 01, 2023 at 09:27:29AM +0200, Sakari Ailus wrote: > Add fwnode_name_eq() to implement the functionality of of_node_name_eq() > on fwnode property API. The same convention of ending the comparison at > '@' (besides '\0') is applied on also both ACPI and swnode. The function > is intended for comparing unit address-less node names on DT and firmware > or swnodes compliant with DT bindings. > > Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Tested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com> > --- > since v1: > > - Use str_has_prefix(). > > - Fix kernel-doc for @fwnode argument. > > - Use size_t type for len. > > The patch is untested. Not anymore :-) Which tree will this get merged through ? The thp7312 driver depends on this patch, and I would like to get it in v6.8. If we can merge this through the media tree it could be the simplest option, otherwise I'll need a stable branch based on v6.7-rc1. > drivers/base/property.c | 28 ++++++++++++++++++++++++++++ > include/linux/property.h | 1 + > 2 files changed, 29 insertions(+) > > diff --git a/drivers/base/property.c b/drivers/base/property.c > index 8c40abed7852..89a6e0833356 100644 > --- a/drivers/base/property.c > +++ b/drivers/base/property.c > @@ -594,6 +594,34 @@ const char *fwnode_get_name_prefix(const struct fwnode_handle *fwnode) > return fwnode_call_ptr_op(fwnode, get_name_prefix); > } > > +/** > + * fwnode_name_eq - Return true if node name is equal > + * @fwnode: The firmware node > + * @name: The name to which to compare the node name > + * > + * Compare the name provided as an argument to the name of the node, stopping > + * the comparison to either '\0' or '@' character, whichever comes first. This > + * function is generally used for comparing node names while ignoring the > + * possible unit address of the node. > + * > + * Return: true if the node name matches with the name provided in the @name > + * argument, false otherwise. > + */ > +bool fwnode_name_eq(const struct fwnode_handle *fwnode, const char *name) > +{ > + const char *node_name; > + size_t len; > + > + node_name = fwnode_get_name(fwnode); > + if (!node_name) > + return false; > + > + len = strchrnul(node_name, '@') - node_name; > + > + return str_has_prefix(node_name, name) == len; > +} > +EXPORT_SYMBOL_GPL(fwnode_name_eq); > + > /** > * fwnode_get_parent - Return parent firwmare node > * @fwnode: Firmware whose parent is retrieved > diff --git a/include/linux/property.h b/include/linux/property.h > index 083a1f41364b..096ade186601 100644 > --- a/include/linux/property.h > +++ b/include/linux/property.h > @@ -108,6 +108,7 @@ struct fwnode_handle *fwnode_find_reference(const struct fwnode_handle *fwnode, > > const char *fwnode_get_name(const struct fwnode_handle *fwnode); > const char *fwnode_get_name_prefix(const struct fwnode_handle *fwnode); > +bool fwnode_name_eq(const struct fwnode_handle *fwnode, const char *name); > > struct fwnode_handle *fwnode_get_parent(const struct fwnode_handle *fwnode); > struct fwnode_handle *fwnode_get_next_parent(struct fwnode_handle *fwnode);
On Wed, Nov 01, 2023 at 10:00:15AM +0000, Sakari Ailus wrote: > On Wed, Nov 01, 2023 at 11:50:00AM +0200, Andy Shevchenko wrote: > > On Wed, Nov 01, 2023 at 09:27:29AM +0200, Sakari Ailus wrote: ... > > > - Use size_t type for len. > > > > Strictly speaking it should be ptrdiff_t according to the code. > > It'll be compared to size_t right after. ptrdiff_t is signed and we know we > have a positive number here so size_t seems like a better choice. Yeah, this is downsides of C language and typing system. But diff between pointers returns ptrdiff_t result. ... > > > + * the comparison to either '\0' or '@' character > > > > NUL is a human-readable substitution to less parsable '\0'. > > I can send v3 with that but I'll wait a bit if there are more comments. Why not? You got tags, so I'll give mine with this being addressed.
diff --git a/drivers/base/property.c b/drivers/base/property.c index 8c40abed7852..89a6e0833356 100644 --- a/drivers/base/property.c +++ b/drivers/base/property.c @@ -594,6 +594,34 @@ const char *fwnode_get_name_prefix(const struct fwnode_handle *fwnode) return fwnode_call_ptr_op(fwnode, get_name_prefix); } +/** + * fwnode_name_eq - Return true if node name is equal + * @fwnode: The firmware node + * @name: The name to which to compare the node name + * + * Compare the name provided as an argument to the name of the node, stopping + * the comparison to either '\0' or '@' character, whichever comes first. This + * function is generally used for comparing node names while ignoring the + * possible unit address of the node. + * + * Return: true if the node name matches with the name provided in the @name + * argument, false otherwise. + */ +bool fwnode_name_eq(const struct fwnode_handle *fwnode, const char *name) +{ + const char *node_name; + size_t len; + + node_name = fwnode_get_name(fwnode); + if (!node_name) + return false; + + len = strchrnul(node_name, '@') - node_name; + + return str_has_prefix(node_name, name) == len; +} +EXPORT_SYMBOL_GPL(fwnode_name_eq); + /** * fwnode_get_parent - Return parent firwmare node * @fwnode: Firmware whose parent is retrieved diff --git a/include/linux/property.h b/include/linux/property.h index 083a1f41364b..096ade186601 100644 --- a/include/linux/property.h +++ b/include/linux/property.h @@ -108,6 +108,7 @@ struct fwnode_handle *fwnode_find_reference(const struct fwnode_handle *fwnode, const char *fwnode_get_name(const struct fwnode_handle *fwnode); const char *fwnode_get_name_prefix(const struct fwnode_handle *fwnode); +bool fwnode_name_eq(const struct fwnode_handle *fwnode, const char *name); struct fwnode_handle *fwnode_get_parent(const struct fwnode_handle *fwnode); struct fwnode_handle *fwnode_get_next_parent(struct fwnode_handle *fwnode);
Add fwnode_name_eq() to implement the functionality of of_node_name_eq() on fwnode property API. The same convention of ending the comparison at '@' (besides '\0') is applied on also both ACPI and swnode. The function is intended for comparing unit address-less node names on DT and firmware or swnodes compliant with DT bindings. Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> --- since v1: - Use str_has_prefix(). - Fix kernel-doc for @fwnode argument. - Use size_t type for len. The patch is untested. drivers/base/property.c | 28 ++++++++++++++++++++++++++++ include/linux/property.h | 1 + 2 files changed, 29 insertions(+)