diff options
author | Bastien Montagne <bastien@blender.org> | 2022-06-09 15:58:35 +0300 |
---|---|---|
committer | Bastien Montagne <bastien@blender.org> | 2022-06-09 16:17:20 +0300 |
commit | 14d1ad8dd8380dab783e6517c437a90cbf73c398 (patch) | |
tree | 95300594d4c2276cca1d635eaf633a37ed3c078b /source/blender | |
parent | 6d726192be73856542dec010d149931bc806e045 (diff) |
FCurve: optimize search from an RNA path + index.
By checking the index value first instead of a full fledge string
comparision in `BKE_fcurve_find`, we can make that code significatly
faster (from about 10% in a Heist production file to over 45% in a
heavily animated test file).
While this code was already very fast (a few microseconds per call
typically), it gets called a lot from the UI (several hundreds of time
per refresh), among other things.
NOTE: the `UNLIKELY` hint is responsible for 25% to 30% of the
speed improvement.
Diffstat (limited to 'source/blender')
-rw-r--r-- | source/blender/blenkernel/intern/fcurve.c | 9 |
1 files changed, 4 insertions, 5 deletions
diff --git a/source/blender/blenkernel/intern/fcurve.c b/source/blender/blenkernel/intern/fcurve.c index 48ba6a851d2..0203620df84 100644 --- a/source/blender/blenkernel/intern/fcurve.c +++ b/source/blender/blenkernel/intern/fcurve.c @@ -256,12 +256,11 @@ FCurve *BKE_fcurve_find(ListBase *list, const char rna_path[], const int array_i /* Check paths of curves, then array indices... */ for (fcu = list->first; fcu; fcu = fcu->next) { + /* Check indices first, much cheaper than a string comparison. */ /* Simple string-compare (this assumes that they have the same root...) */ - if (fcu->rna_path && STREQ(fcu->rna_path, rna_path)) { - /* Now check indices. */ - if (fcu->array_index == array_index) { - return fcu; - } + if (UNLIKELY(fcu->array_index == array_index && fcu->rna_path && + fcu->rna_path[0] == rna_path[0] && STREQ(fcu->rna_path, rna_path))) { + return fcu; } } |