On Mon, Oct 28, 2024 at 04:49:40AM +0100, Ole Schuerks wrote:
> +/**
> + * list_size - counts the number of entries in a list
> + * @head: the list whose entries are counted
> + */
> +static inline size_t list_size(const struct list_head *head)
> +{
> + size_t ret = 0;
> +
> + for (struct list_head *curr = head->next; curr != head;
> + curr = curr->next)
> + ++ret;
> +
> + return ret;
What's wrong with list_count_nodes()?
> @@ -310,6 +325,40 @@ static inline int list_empty(const struct list_head *head)
> !list_entry_is_head(pos, head, member); \
> pos = n, n = list_next_entry(n, member))
>
> +/**
> + * list_for_each_entry_from - iterate over list of given type starting at a given node
> + * @pos: the type * to use as a loop cursor.
> + * @start: the node to start iterating at
> + * @head: the head for your list.
> + * @member: the name of the list_head within the struct.
> + */
> +#define list_for_each_entry_from(pos, start, head, member) \
> + for (pos = list_entry(start, typeof(*pos), member); \
> + !list_entry_is_head(pos, head, member); \
> + pos = list_next_entry(pos, member))
list_for_each_entry_from() exists on my tree on linux-next added
through commit e229c2fb3370a ("[LIST]: Introduce
list_for_each_entry_from") since v2.6.17, so since 2006.
> +/**
> + * list_at_index - retrieve the entry at index i in O(n)
> + * @i: index of entry to retrieve.
> + * @head: the head for your list.
> + * @type: the type of the struct the entries are embedded in.
> + * @member: the name of the list_head within the struct.
> + */
> +#define list_at_index(i, head, type, member) \
> + ({ \
> + type *__pos; \
> + size_t __counter = 0; \
> + list_for_each_entry(__pos, head, member) { \
> + if (__counter++ == i) \
> + break; \
> + if (__pos->member.next == head) { \
> + __pos = NULL; \
> + break; \
> + } \
> + } \
> + __pos; \
> + })
> +
Seems like the only thing being added. I'd just keep thsi internal
to your code for now, and later if we really want we can add this
as a generic helper. If you really want this on list.h I'd recommend
to find a few users of this already in kernel and generalize the common
thing to use it first and add it that way. Then you separate *that*
patches series from this one, and address that first, and address
the addition then as a seprate patch series.
If you keep it to the tool itself, just change the namespace to be not
be as generically named of course.
Luis
