tmaq
/
tornavis
1
0
Fork 0
Code Issues 2 Pull Requests 29 Packages Projects 24 Releases Wiki Activity

BLI_array_store: improve hash accumulation for each memory chunk 

The method of accumulating values to create a hash for each chunk has
been improved for ~16% better distribution of the resulting hashes.

Improve performance of array de-duplication, see: #105046.
This commit is contained in:
Campbell Barton 2023-03-11 11:57:13 +11:00
parent 7a5cb3b470
commit 16968ccec7
1 changed files with 11 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
source/blender/blenlib/intern/array_store.c
View File

@ -833,6 +833,14 @@ static void hash_array_from_cref(const BArrayInfo *info,
BLI_assert(i == hash_array_len);
}
BLI_INLINE void hash_accum_impl(hash_key *hash_array, const size_t i_dst, const size_t i_ahead)
{
/* Tested to give good results when accumulating unique values from an array of booleans.
* (least unused cells in the `BTableRef **table`). */
BLI_assert(i_dst < i_ahead);
hash_array[i_dst] += ((hash_array[i_ahead] << 3) ^ (hash_array[i_dst] >> 1));
}
static void hash_accum(hash_key *hash_array, const size_t hash_array_len, size_t iter_steps)
{
/* _very_ unlikely, can happen if you select a chunk-size of 1 for example. */
@ -843,8 +851,8 @@ static void hash_accum(hash_key *hash_array, const size_t hash_array_len, size_t
const size_t hash_array_search_len = hash_array_len - iter_steps;
while (iter_steps != 0) {
const size_t hash_offset = iter_steps;
for (uint i = 0; i < hash_array_search_len; i++) {
hash_array[i] += (hash_array[i + hash_offset]) * ((hash_array[i] & 0xff) + 1);
for (size_t i = 0; i < hash_array_search_len; i++) {
hash_accum_impl(hash_array, i, i + hash_offset);
}
iter_steps -= 1;
}
@ -869,7 +877,7 @@ static void hash_accum_single(hash_key *hash_array, const size_t hash_array_len,
const size_t hash_array_search_len = hash_array_len - iter_steps_sub;
const size_t hash_offset = iter_steps;
for (uint i = 0; i < hash_array_search_len; i++) {
hash_array[i] += (hash_array[i + hash_offset]) * ((hash_array[i] & 0xff) + 1);
hash_accum_impl(hash_array, i, i + hash_offset);
}
iter_steps -= 1;
iter_steps_sub += iter_steps;