hashmap: introduce hashmap_reserve()

With the current hashmap implementation that uses chaining, placing a
reservation can serve two purposes:
 - To optimize putting of entries if the number of entries to put is
   known. The reservation allocates buckets, so later resizing can be
   avoided.
 - To avoid having very long bucket chains after using
   hashmap_move(_one).

In an alternative hashmap implementation it will serve an additional
purpose:
 - To guarantee a subsequent hashmap_move(_one) will not fail with
   -ENOMEM (this never happens in the current implementation).

Signed-off-by: Anthony G. Basile <blueness@gentoo.org>

4 files changed, 33 insertions, 6 deletions

diff --git a/src/shared/hashmap.c b/src/shared/hashmap.c
index e965167258..fa4edf11c6 100644
--- a/src/shared/hashmap.c
+++ b/src/shared/hashmap.c
@@ -276,18 +276,26 @@ static struct hashmap_entry *hash_scan(Hashmap *h, unsigned hash, const void *ke
         return NULL;
 }
 
-static int resize_buckets(Hashmap *h) {
+static int resize_buckets(Hashmap *h, unsigned entries_add) {
         struct hashmap_entry **n, *i;
-        unsigned m;
+        unsigned m, new_n_entries, new_n_buckets;
         uint8_t nkey[HASH_KEY_SIZE];
 
         assert(h);
 
-        if (_likely_(h->n_entries*4 < h->n_buckets*3))
+        new_n_entries = h->n_entries + entries_add;
+
+        /* overflow? */
+        if (_unlikely_(new_n_entries < entries_add || new_n_entries > UINT_MAX / 4))
+                return -ENOMEM;
+
+        new_n_buckets = new_n_entries * 4 / 3;
+
+        if (_likely_(new_n_buckets <= h->n_buckets))
                 return 0;
 
-        /* Increase by four */
-        m = (h->n_entries+1)*4-1;
+        /* Increase by four at least */
+        m = MAX((h->n_entries+1)*4-1, new_n_buckets);
 
         /* If we hit OOM we simply risk packed hashmaps... */
         n = new0(struct hashmap_entry*, m);
@@ -339,7 +347,7 @@ static int __hashmap_put(Hashmap *h, const void *key, void *value, unsigned hash
 
         struct hashmap_entry *e;
 
-        if (resize_buckets(h) > 0)
+        if (resize_buckets(h, 1) > 0)
                 hash = bucket_hash(h, key);
 
         if (h->from_pool)
@@ -476,6 +484,18 @@ unsigned hashmap_size(Hashmap *h) {
         return h->n_entries;
 }
 
+int hashmap_reserve(Hashmap *h, unsigned entries_add) {
+        int r;
+
+        assert(h);
+
+        r = resize_buckets(h, entries_add);
+        if (r < 0)
+                return r;
+
+        return 0;
+}
+
 char **hashmap_get_strv(Hashmap *h) {
         char **sv;
         Iterator it;
diff --git a/src/shared/hashmap.h b/src/shared/hashmap.h
index 53b04eb033..22bda37437 100644
--- a/src/shared/hashmap.h
+++ b/src/shared/hashmap.h
@@ -65,6 +65,7 @@ int hashmap_put(Hashmap *h, const void *key, void *value);
 void *hashmap_get(Hashmap *h, const void *key);
 void *hashmap_get2(Hashmap *h, const void *key, void **rkey);
 bool hashmap_contains(Hashmap *h, const void *key);
+int hashmap_reserve(Hashmap *h, unsigned entries_add);
 
 unsigned hashmap_size(Hashmap *h) _pure_;
 
diff --git a/src/shared/set.c b/src/shared/set.c
index 4b8c76feca..087d874598 100644
--- a/src/shared/set.c
+++ b/src/shared/set.c
@@ -50,3 +50,7 @@ bool set_contains(Set *s, void *value) {
 void *set_iterate(Set *s, Iterator *i) {
         return hashmap_iterate(MAKE_HASHMAP(s), i, NULL);
 }
+
+int set_reserve(Set *s, unsigned entries_add) {
+        return hashmap_reserve(MAKE_HASHMAP(s), entries_add);
+}
diff --git a/src/shared/set.h b/src/shared/set.h
index f1111d1bc8..ae3090a09f 100644
--- a/src/shared/set.h
+++ b/src/shared/set.h
@@ -37,6 +37,8 @@ int set_put(Set *s, void *value);
 void *set_get(Set *s, void *value);
 bool set_contains(Set *s, void *value);
 
+int set_reserve(Set *s, unsigned entries_add);
+
 void *set_iterate(Set *s, Iterator *i);
 
 DEFINE_TRIVIAL_CLEANUP_FUNC(Set*, set_free);