name-hash.c: remove cache entries instead of marking them CE_UNHASHED
The new hashmap implementation supports remove, so really remove unused cache entries from the name hashmap instead of just marking them. The CE_UNHASHED flag and CE_STATE_MASK are no longer needed. Keep the CE_HASHED flag to prevent adding entries twice. Signed-off-by: Karsten Blees <blees@dcon.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Karsten Blees committed
Nov 14, 2013 at 20:22 UTC
419a597f6408d5b5cffe3e278258f58b47d15ad3
4 files changed
+26
-30
cache.h
+2
-4
@@ -160,7 +160,6 @@ struct cache_entry {
160
#define CE_ADDED (1 << 19)
161
162
#define CE_HASHED (1 << 20)
163
-#define CE_UNHASHED (1 << 21)
163
#define CE_WT_REMOVE (1 << 22) /* remove in work directory */
164
#define CE_CONFLICTED (1 << 23)
165
@@ -196,11 +195,10 @@ struct pathspec;
195
* Copy the sha1 and stat state of a cache entry from one to
196
* another. But we never change the name, or the hash state!
197
*/
199
-#define CE_STATE_MASK (CE_HASHED | CE_UNHASHED)
198
static inline void copy_cache_entry(struct cache_entry *dst,
199
const struct cache_entry *src)
200
{
203
- unsigned int state = dst->ce_flags & CE_STATE_MASK;
201
+ unsigned int state = dst->ce_flags & CE_HASHED;
202
203
/* Don't copy hash chain and name */
204
memcpy(&dst->ce_stat_data, &src->ce_stat_data,
@@ -208,7 +206,7 @@ static inline void copy_cache_entry(struct cache_entry *dst,
206
offsetof(struct cache_entry, ce_stat_data));
207
208
/* Restore the hash state */
211
- dst->ce_flags = (dst->ce_flags & ~CE_STATE_MASK) | state;
209
+ dst->ce_flags = (dst->ce_flags & ~CE_HASHED) | state;
210
}
211
212
static inline unsigned create_ce_flags(unsigned stage)
name-hash.c
+22
-24
@@ -106,17 +106,29 @@ static void hash_index_entry(struct index_state *istate, struct cache_entry *ce)
106
hashmap_entry_init(ce, memihash(ce->name, ce_namelen(ce)));
107
hashmap_add(&istate->name_hash, ce);
108
109
- if (ignore_case && !(ce->ce_flags & CE_UNHASHED))
109
+ if (ignore_case)
110
add_dir_entry(istate, ce);
111
}
112
113
+static int cache_entry_cmp(const struct cache_entry *ce1,
114
+ const struct cache_entry *ce2, const void *remove)
115
+{
116
+ /*
117
+ * For remove_name_hash, find the exact entry (pointer equality); for
118
+ * index_name_exists, find all entries with matching hash code and
119
+ * decide whether the entry matches in same_name.
120
+ */
121
+ return remove ? !(ce1 == ce2) : 0;
122
+}
123
+
124
static void lazy_init_name_hash(struct index_state *istate)
125
{
126
int nr;
127
128
if (istate->name_hash_initialized)
129
return;
119
- hashmap_init(&istate->name_hash, NULL, istate->cache_nr);
130
+ hashmap_init(&istate->name_hash, (hashmap_cmp_fn) cache_entry_cmp,
131
+ istate->cache_nr);
132
hashmap_init(&istate->dir_hash, (hashmap_cmp_fn) dir_entry_cmp, 0);
133
for (nr = 0; nr < istate->cache_nr; nr++)
134
hash_index_entry(istate, istate->cache[nr]);
@@ -125,31 +137,19 @@ static void lazy_init_name_hash(struct index_state *istate)
137
138
void add_name_hash(struct index_state *istate, struct cache_entry *ce)
139
{
128
- /* if already hashed, add reference to directory entries */
129
- if (ignore_case && (ce->ce_flags & CE_STATE_MASK) == CE_STATE_MASK)
130
- add_dir_entry(istate, ce);
131
-
132
- ce->ce_flags &= ~CE_UNHASHED;
140
if (istate->name_hash_initialized)
141
hash_index_entry(istate, ce);
142
}
143
137
-/*
138
- * We don't actually *remove* it, we can just mark it invalid so that
139
- * we won't find it in lookups.
140
- *
141
- * Not only would we have to search the lists (simple enough), but
142
- * we'd also have to rehash other hash buckets in case this makes the
143
- * hash bucket empty (common). So it's much better to just mark
144
- * it.
145
- */
144
void remove_name_hash(struct index_state *istate, struct cache_entry *ce)
145
{
148
- /* if already hashed, release reference to directory entries */
149
- if (ignore_case && (ce->ce_flags & CE_STATE_MASK) == CE_HASHED)
150
- remove_dir_entry(istate, ce);
146
+ if (!istate->name_hash_initialized || !(ce->ce_flags & CE_HASHED))
147
+ return;
148
+ ce->ce_flags &= ~CE_HASHED;
149
+ hashmap_remove(&istate->name_hash, ce, ce);
150
152
- ce->ce_flags |= CE_UNHASHED;
151
+ if (ignore_case)
152
+ remove_dir_entry(istate, ce);
153
}
154
155
static int slow_same_name(const char *name1, int len1, const char *name2, int len2)
@@ -220,10 +220,8 @@ struct cache_entry *index_file_exists(struct index_state *istate, const char *na
220
hashmap_entry_init(&key, memihash(name, namelen));
221
ce = hashmap_get(&istate->name_hash, &key, NULL);
222
while (ce) {
223
- if (!(ce->ce_flags & CE_UNHASHED)) {
224
- if (same_name(ce, name, namelen, icase))
225
- return ce;
226
- }
223
+ if (same_name(ce, name, namelen, icase))
224
+ return ce;
225
ce = hashmap_get_next(&istate->name_hash, ce);
226
}
227
return NULL;
read-cache.c
+1
-1
@@ -58,7 +58,7 @@ void rename_index_entry_at(struct index_state *istate, int nr, const char *new_n
58
59
new = xmalloc(cache_entry_size(namelen));
60
copy_cache_entry(new, old);
61
- new->ce_flags &= ~CE_STATE_MASK;
61
+ new->ce_flags &= ~CE_HASHED;
62
new->ce_namelen = namelen;
63
memcpy(new->name, new_name, namelen + 1);
64
unpack-trees.c
+1
-1
@@ -105,7 +105,7 @@ void setup_unpack_trees_porcelain(struct unpack_trees_options *opts,
105
static void do_add_entry(struct unpack_trees_options *o, struct cache_entry *ce,
106
unsigned int set, unsigned int clear)
107
{
108
- clear |= CE_HASHED | CE_UNHASHED;
108
+ clear |= CE_HASHED;
109
110
if (set & CE_REMOVE)
111
set |= CE_WT_REMOVE;