rcu: Introduce hlist_nulls variant of hlist

hlist uses NULL value to finish a chain.

hlist_nulls variant use the low order bit set to 1 to signal an end-of-list marker.

This allows to store many different end markers, so that some RCU lockless
algos (used in TCP/UDP stack for example) can save some memory barriers in
fast paths.

Two new files are added :

include/linux/list_nulls.h
  - mimics hlist part of include/linux/list.h, derived to hlist_nulls variant

include/linux/rculist_nulls.h
  - mimics hlist part of include/linux/rculist.h, derived to hlist_nulls variant

   Only four helpers are declared for the moment :

     hlist_nulls_del_init_rcu(), hlist_nulls_del_rcu(),
     hlist_nulls_add_head_rcu() and hlist_nulls_for_each_entry_rcu()

prefetches() were removed, since an end of list is not anymore NULL value.
prefetches() could trigger useless (and possibly dangerous) memory transactions.

Example of use (extracted from __udp4_lib_lookup())

	struct sock *sk, *result;
        struct hlist_nulls_node *node;
        unsigned short hnum = ntohs(dport);
        unsigned int hash = udp_hashfn(net, hnum);
        struct udp_hslot *hslot = &udptable->hash[hash];
        int score, badness;

        rcu_read_lock();
begin:
        result = NULL;
        badness = -1;
        sk_nulls_for_each_rcu(sk, node, &hslot->head) {
                score = compute_score(sk, net, saddr, hnum, sport,
                                      daddr, dport, dif);
                if (score > badness) {
                        result = sk;
                        badness = score;
                }
        }
        /*
         * if the nulls value we got at the end of this lookup is
         * not the expected one, we must restart lookup.
         * We probably met an item that was moved to another chain.
         */
        if (get_nulls_value(node) != hash)
                goto begin;

        if (result) {
                if (unlikely(!atomic_inc_not_zero(&result->sk_refcnt)))
                        result = NULL;
                else if (unlikely(compute_score(result, net, saddr, hnum, sport,
                                  daddr, dport, dif) < badness)) {
                        sock_put(result);
                        goto begin;
                }
        }
        rcu_read_unlock();
        return result;

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: David S. Miller <davem@davemloft.net>

1 files changed, 94 insertions, 0 deletions

diff --git a/include/linux/list_nulls.h b/include/linux/list_nulls.h
new file mode 100644
index 000000000000..93150ecf3ea4
--- /dev/null
+++ b/include/linux/list_nulls.h
@@ -0,0 +1,94 @@
+#ifndef _LINUX_LIST_NULLS_H
+#define _LINUX_LIST_NULLS_H
+
+/*
+ * Special version of lists, where end of list is not a NULL pointer,
+ * but a 'nulls' marker, which can have many different values.
+ * (up to 2^31 different values guaranteed on all platforms)
+ *
+ * In the standard hlist, termination of a list is the NULL pointer.
+ * In this special 'nulls' variant, we use the fact that objects stored in
+ * a list are aligned on a word (4 or 8 bytes alignment).
+ * We therefore use the last significant bit of 'ptr' :
+ * Set to 1 : This is a 'nulls' end-of-list marker (ptr >> 1)
+ * Set to 0 : This is a pointer to some object (ptr)
+ */
+
+struct hlist_nulls_head {
+        struct hlist_nulls_node *first;
+};
+
+struct hlist_nulls_node {
+        struct hlist_nulls_node *next, **pprev;
+};
+#define INIT_HLIST_NULLS_HEAD(ptr, nulls) \
+        ((ptr)->first = (struct hlist_nulls_node *) (1UL | (((long)nulls) << 1)))
+
+#define hlist_nulls_entry(ptr, type, member) container_of(ptr,type,member)
+/**
+ * ptr_is_a_nulls - Test if a ptr is a nulls
+ * @ptr: ptr to be tested
+ *
+ */
+static inline int is_a_nulls(const struct hlist_nulls_node *ptr)
+{
+        return ((unsigned long)ptr & 1);
+}
+
+/**
+ * get_nulls_value - Get the 'nulls' value of the end of chain
+ * @ptr: end of chain
+ *
+ * Should be called only if is_a_nulls(ptr);
+ */
+static inline unsigned long get_nulls_value(const struct hlist_nulls_node *ptr)
+{
+        return ((unsigned long)ptr) >> 1;
+}
+
+static inline int hlist_nulls_unhashed(const struct hlist_nulls_node *h)
+{
+        return !h->pprev;
+}
+
+static inline int hlist_nulls_empty(const struct hlist_nulls_head *h)
+{
+        return is_a_nulls(h->first);
+}
+
+static inline void __hlist_nulls_del(struct hlist_nulls_node *n)
+{
+        struct hlist_nulls_node *next = n->next;
+        struct hlist_nulls_node **pprev = n->pprev;
+        *pprev = next;
+        if (!is_a_nulls(next))
+                next->pprev = pprev;
+}
+
+/**
+ * hlist_nulls_for_each_entry   - iterate over list of given type
+ * @tpos:       the type * to use as a loop cursor.
+ * @pos:        the &struct hlist_node to use as a loop cursor.
+ * @head:       the head for your list.
+ * @member:     the name of the hlist_node within the struct.
+ *
+ */
+#define hlist_nulls_for_each_entry(tpos, pos, head, member)                    \
+        for (pos = (head)->first;                                              \
+             (!is_a_nulls(pos)) &&                                             \
+                ({ tpos = hlist_nulls_entry(pos, typeof(*tpos), member); 1;}); \
+             pos = pos->next)
+
+/**
+ * hlist_nulls_for_each_entry_from - iterate over a hlist continuing from current point
+ * @tpos:       the type * to use as a loop cursor.
+ * @pos:        the &struct hlist_node to use as a loop cursor.
+ * @member:     the name of the hlist_node within the struct.
+ *
+ */
+#define hlist_nulls_for_each_entry_from(tpos, pos, member)      \
+        for (; (!is_a_nulls(pos)) &&                            \
+                ({ tpos = hlist_nulls_entry(pos, typeof(*tpos), member); 1;}); \
+             pos = pos->next)
+
+#endif