-
Committer:
John Arbash Meinel
-
Date:
2008-07-25 03:22:23 UTC
-
mto:
(0.17.20 trunk)
-
mto:
This revision was merged to the branch mainline in
revision
4280.
-
Revision ID:
john@arbash-meinel.com-20080725032223-uz2pbbqz5ez2t3mi
It turns out that appending a short set of lines every time was killing performance in realloc.
So fudge a bit and allocate memory up to the next power of 2.
This may be a bit overzealous, but we do need something to prevent a realloc for every extend.