Hi, On Tue, Oct 22, 2013 at 01:40:18PM +0200, Gert Doering wrote: > I could use some ideas on how to debug this further - if valgrind isn't > complaining, normally our memory management should be OK, but why is > VSZ/RSZ still growing, then...?
Arne brought up a good point. If we "lose" the memory inside the gc_*
functions (as in: use the same gc block for a very long time, like
"while OpenVPN runs", but *do* free it at the end), valgrind might not
be able to see it - after all, we nicely cleaned up later - but we'll
still grow out of bounds at runtime.
Now, since 2.1 is supposedly not doing this, but 2.2.2 *is*, and the
difference between those versions is not all that big, I really wonder
where and how this got introduced. Maybe time to do some git bisect...
gert
--
USENET is *not* the non-clickable part of WWW!
//www.muc.de/~gert/
Gert Doering - Munich, Germany g...@greenie.muc.de
fax: +49-89-35655025 g...@net.informatik.tu-muenchen.de
pgpm10vY4zFRk.pgp
Description: PGP signature
