Cache HvFILL() for larger hashes, and update on insertion/deletion.

This avoids HvFILL() being O(n) for large n on large hashes, but also avoids
storing the value of HvFILL() in smaller hashes (ie a memory overhead on
every single object built using a hash.)

1 files changed, 1 insertions, 0 deletions

diff --git a/sv.c b/sv.c
index ba09305104..ee5a9d6a7c 100644
--- a/sv.c
+++ b/sv.c
@@ -12349,6 +12349,7 @@ S_sv_dup_common(pTHX_ const SV *const sstr, CLONE_PARAMS *const param)
 			}
 			daux->xhv_name_count = saux->xhv_name_count;
 
+			daux->xhv_fill_lazy = saux->xhv_fill_lazy;
 			daux->xhv_riter = saux->xhv_riter;
 			daux->xhv_eiter = saux->xhv_eiter
 			    ? he_dup(saux->xhv_eiter,