diff options
author | Daniel Borkmann <daniel@iogearbox.net> | 2015-03-12 10:28:40 -0400 |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2015-03-12 23:02:30 -0400 |
commit | a5b6846f9e1a080493210013385c28faecee36f0 (patch) | |
tree | 6b068157f7e2efb5d71978d3a63d14754bcd6a7c /lib | |
parent | 9497df88ab5567daa001829051c5f87161a81ff0 (diff) |
rhashtable: kill ht->shift atomic operations
Commit c0c09bfdc415 ("rhashtable: avoid unnecessary wakeup for worker
queue") changed ht->shift to be atomic, which is actually unnecessary.
Instead of leaving the current shift in the core rhashtable structure,
it can be cached inside the individual bucket tables.
There, it will only be initialized once during a new table allocation
in the shrink/expansion slow path, and from then onward it stays immutable
for the rest of the bucket table liftime.
That allows shift to be non-atomic. The patch also moves hash_rnd
management into the table setup. The rhashtable structure now consumes
3 instead of 4 cachelines.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Cc: Ying Xue <ying.xue@windriver.com>
Acked-by: Thomas Graf <tgraf@suug.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'lib')
-rw-r--r-- | lib/rhashtable.c | 55 |
1 files changed, 25 insertions, 30 deletions
diff --git a/lib/rhashtable.c b/lib/rhashtable.c index 68210cc2bab8..adea791ea3ab 100644 --- a/lib/rhashtable.c +++ b/lib/rhashtable.c | |||
@@ -147,7 +147,7 @@ static void bucket_table_free(const struct bucket_table *tbl) | |||
147 | } | 147 | } |
148 | 148 | ||
149 | static struct bucket_table *bucket_table_alloc(struct rhashtable *ht, | 149 | static struct bucket_table *bucket_table_alloc(struct rhashtable *ht, |
150 | size_t nbuckets) | 150 | size_t nbuckets, u32 hash_rnd) |
151 | { | 151 | { |
152 | struct bucket_table *tbl = NULL; | 152 | struct bucket_table *tbl = NULL; |
153 | size_t size; | 153 | size_t size; |
@@ -162,6 +162,8 @@ static struct bucket_table *bucket_table_alloc(struct rhashtable *ht, | |||
162 | return NULL; | 162 | return NULL; |
163 | 163 | ||
164 | tbl->size = nbuckets; | 164 | tbl->size = nbuckets; |
165 | tbl->shift = ilog2(nbuckets); | ||
166 | tbl->hash_rnd = hash_rnd; | ||
165 | 167 | ||
166 | if (alloc_bucket_locks(ht, tbl) < 0) { | 168 | if (alloc_bucket_locks(ht, tbl) < 0) { |
167 | bucket_table_free(tbl); | 169 | bucket_table_free(tbl); |
@@ -177,25 +179,27 @@ static struct bucket_table *bucket_table_alloc(struct rhashtable *ht, | |||
177 | /** | 179 | /** |
178 | * rht_grow_above_75 - returns true if nelems > 0.75 * table-size | 180 | * rht_grow_above_75 - returns true if nelems > 0.75 * table-size |
179 | * @ht: hash table | 181 | * @ht: hash table |
180 | * @new_size: new table size | 182 | * @tbl: current table |
181 | */ | 183 | */ |
182 | static bool rht_grow_above_75(const struct rhashtable *ht, size_t new_size) | 184 | static bool rht_grow_above_75(const struct rhashtable *ht, |
185 | const struct bucket_table *tbl) | ||
183 | { | 186 | { |
184 | /* Expand table when exceeding 75% load */ | 187 | /* Expand table when exceeding 75% load */ |
185 | return atomic_read(&ht->nelems) > (new_size / 4 * 3) && | 188 | return atomic_read(&ht->nelems) > (tbl->size / 4 * 3) && |
186 | (!ht->p.max_shift || atomic_read(&ht->shift) < ht->p.max_shift); | 189 | (!ht->p.max_shift || tbl->shift < ht->p.max_shift); |
187 | } | 190 | } |
188 | 191 | ||
189 | /** | 192 | /** |
190 | * rht_shrink_below_30 - returns true if nelems < 0.3 * table-size | 193 | * rht_shrink_below_30 - returns true if nelems < 0.3 * table-size |
191 | * @ht: hash table | 194 | * @ht: hash table |
192 | * @new_size: new table size | 195 | * @tbl: current table |
193 | */ | 196 | */ |
194 | static bool rht_shrink_below_30(const struct rhashtable *ht, size_t new_size) | 197 | static bool rht_shrink_below_30(const struct rhashtable *ht, |
198 | const struct bucket_table *tbl) | ||
195 | { | 199 | { |
196 | /* Shrink table beneath 30% load */ | 200 | /* Shrink table beneath 30% load */ |
197 | return atomic_read(&ht->nelems) < (new_size * 3 / 10) && | 201 | return atomic_read(&ht->nelems) < (tbl->size * 3 / 10) && |
198 | (atomic_read(&ht->shift) > ht->p.min_shift); | 202 | tbl->shift > ht->p.min_shift; |
199 | } | 203 | } |
200 | 204 | ||
201 | static int rhashtable_rehash_one(struct rhashtable *ht, unsigned old_hash) | 205 | static int rhashtable_rehash_one(struct rhashtable *ht, unsigned old_hash) |
@@ -310,16 +314,11 @@ int rhashtable_expand(struct rhashtable *ht) | |||
310 | 314 | ||
311 | ASSERT_RHT_MUTEX(ht); | 315 | ASSERT_RHT_MUTEX(ht); |
312 | 316 | ||
313 | new_tbl = bucket_table_alloc(ht, old_tbl->size * 2); | 317 | new_tbl = bucket_table_alloc(ht, old_tbl->size * 2, old_tbl->hash_rnd); |
314 | if (new_tbl == NULL) | 318 | if (new_tbl == NULL) |
315 | return -ENOMEM; | 319 | return -ENOMEM; |
316 | 320 | ||
317 | new_tbl->hash_rnd = old_tbl->hash_rnd; | ||
318 | |||
319 | atomic_inc(&ht->shift); | ||
320 | |||
321 | rhashtable_rehash(ht, new_tbl); | 321 | rhashtable_rehash(ht, new_tbl); |
322 | |||
323 | return 0; | 322 | return 0; |
324 | } | 323 | } |
325 | EXPORT_SYMBOL_GPL(rhashtable_expand); | 324 | EXPORT_SYMBOL_GPL(rhashtable_expand); |
@@ -342,20 +341,15 @@ EXPORT_SYMBOL_GPL(rhashtable_expand); | |||
342 | */ | 341 | */ |
343 | int rhashtable_shrink(struct rhashtable *ht) | 342 | int rhashtable_shrink(struct rhashtable *ht) |
344 | { | 343 | { |
345 | struct bucket_table *new_tbl, *tbl = rht_dereference(ht->tbl, ht); | 344 | struct bucket_table *new_tbl, *old_tbl = rht_dereference(ht->tbl, ht); |
346 | 345 | ||
347 | ASSERT_RHT_MUTEX(ht); | 346 | ASSERT_RHT_MUTEX(ht); |
348 | 347 | ||
349 | new_tbl = bucket_table_alloc(ht, tbl->size / 2); | 348 | new_tbl = bucket_table_alloc(ht, old_tbl->size / 2, old_tbl->hash_rnd); |
350 | if (new_tbl == NULL) | 349 | if (new_tbl == NULL) |
351 | return -ENOMEM; | 350 | return -ENOMEM; |
352 | 351 | ||
353 | new_tbl->hash_rnd = tbl->hash_rnd; | ||
354 | |||
355 | atomic_dec(&ht->shift); | ||
356 | |||
357 | rhashtable_rehash(ht, new_tbl); | 352 | rhashtable_rehash(ht, new_tbl); |
358 | |||
359 | return 0; | 353 | return 0; |
360 | } | 354 | } |
361 | EXPORT_SYMBOL_GPL(rhashtable_shrink); | 355 | EXPORT_SYMBOL_GPL(rhashtable_shrink); |
@@ -376,9 +370,9 @@ static void rht_deferred_worker(struct work_struct *work) | |||
376 | list_for_each_entry(walker, &ht->walkers, list) | 370 | list_for_each_entry(walker, &ht->walkers, list) |
377 | walker->resize = true; | 371 | walker->resize = true; |
378 | 372 | ||
379 | if (rht_grow_above_75(ht, tbl->size)) | 373 | if (rht_grow_above_75(ht, tbl)) |
380 | rhashtable_expand(ht); | 374 | rhashtable_expand(ht); |
381 | else if (rht_shrink_below_30(ht, tbl->size)) | 375 | else if (rht_shrink_below_30(ht, tbl)) |
382 | rhashtable_shrink(ht); | 376 | rhashtable_shrink(ht); |
383 | unlock: | 377 | unlock: |
384 | mutex_unlock(&ht->mutex); | 378 | mutex_unlock(&ht->mutex); |
@@ -431,7 +425,7 @@ static bool __rhashtable_insert(struct rhashtable *ht, struct rhash_head *obj, | |||
431 | rcu_assign_pointer(tbl->buckets[hash], obj); | 425 | rcu_assign_pointer(tbl->buckets[hash], obj); |
432 | 426 | ||
433 | atomic_inc(&ht->nelems); | 427 | atomic_inc(&ht->nelems); |
434 | if (no_resize_running && rht_grow_above_75(ht, tbl->size)) | 428 | if (no_resize_running && rht_grow_above_75(ht, tbl)) |
435 | schedule_work(&ht->run_work); | 429 | schedule_work(&ht->run_work); |
436 | 430 | ||
437 | exit: | 431 | exit: |
@@ -539,7 +533,7 @@ bool rhashtable_remove(struct rhashtable *ht, struct rhash_head *obj) | |||
539 | bool no_resize_running = tbl == old_tbl; | 533 | bool no_resize_running = tbl == old_tbl; |
540 | 534 | ||
541 | atomic_dec(&ht->nelems); | 535 | atomic_dec(&ht->nelems); |
542 | if (no_resize_running && rht_shrink_below_30(ht, tbl->size)) | 536 | if (no_resize_running && rht_shrink_below_30(ht, tbl)) |
543 | schedule_work(&ht->run_work); | 537 | schedule_work(&ht->run_work); |
544 | } | 538 | } |
545 | 539 | ||
@@ -913,6 +907,7 @@ int rhashtable_init(struct rhashtable *ht, struct rhashtable_params *params) | |||
913 | { | 907 | { |
914 | struct bucket_table *tbl; | 908 | struct bucket_table *tbl; |
915 | size_t size; | 909 | size_t size; |
910 | u32 hash_rnd; | ||
916 | 911 | ||
917 | size = HASH_DEFAULT_SIZE; | 912 | size = HASH_DEFAULT_SIZE; |
918 | 913 | ||
@@ -939,14 +934,14 @@ int rhashtable_init(struct rhashtable *ht, struct rhashtable_params *params) | |||
939 | else | 934 | else |
940 | ht->p.locks_mul = BUCKET_LOCKS_PER_CPU; | 935 | ht->p.locks_mul = BUCKET_LOCKS_PER_CPU; |
941 | 936 | ||
942 | tbl = bucket_table_alloc(ht, size); | 937 | get_random_bytes(&hash_rnd, sizeof(hash_rnd)); |
938 | |||
939 | tbl = bucket_table_alloc(ht, size, hash_rnd); | ||
943 | if (tbl == NULL) | 940 | if (tbl == NULL) |
944 | return -ENOMEM; | 941 | return -ENOMEM; |
945 | 942 | ||
946 | get_random_bytes(&tbl->hash_rnd, sizeof(tbl->hash_rnd)); | ||
947 | |||
948 | atomic_set(&ht->nelems, 0); | 943 | atomic_set(&ht->nelems, 0); |
949 | atomic_set(&ht->shift, ilog2(tbl->size)); | 944 | |
950 | RCU_INIT_POINTER(ht->tbl, tbl); | 945 | RCU_INIT_POINTER(ht->tbl, tbl); |
951 | RCU_INIT_POINTER(ht->future_tbl, tbl); | 946 | RCU_INIT_POINTER(ht->future_tbl, tbl); |
952 | 947 | ||