From abf70309eb42674982ea51ed41ff5b39b04fe701 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Viktor=20S=C3=B6derqvist?= Date: Sat, 26 Nov 2022 02:35:18 +0100 Subject: Shrink dict without rehashing (#11540) When we're shrinking the hash table, we don't need to hash the keys. Since the table sizes are powers of two, we can simply mask the bucket index in the larger table to get the bucket index in the smaller table. We avoid loading the keys into memory and save CPU time. --- src/dict.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) (limited to 'src/dict.c') diff --git a/src/dict.c b/src/dict.c index 9975e7872..9ed461d62 100644 --- a/src/dict.c +++ b/src/dict.c @@ -229,7 +229,14 @@ int dictRehash(dict *d, int n) { nextde = de->next; /* Get the index in the new hash table */ - h = dictHashKey(d, de->key) & DICTHT_SIZE_MASK(d->ht_size_exp[1]); + if (d->ht_size_exp[1] > d->ht_size_exp[0]) { + h = dictHashKey(d, de->key) & DICTHT_SIZE_MASK(d->ht_size_exp[1]); + } else { + /* We're shrinking the table. The tables sizes are powers of + * two, so we simply mask the bucket index in the larger table + * to get the bucket index in the smaller table. */ + h = d->rehashidx & DICTHT_SIZE_MASK(d->ht_size_exp[1]); + } de->next = d->ht_table[1][h]; d->ht_table[1][h] = de; d->ht_used[0]--; -- cgit v1.2.1