http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59061
--- Comment #20 from Kostya Serebryany <kcc at gcc dot gnu.org> --- > I our simulation code, it looks like the overhead for leak checking is about > 20%. I haven't done very careful measurements yet, since this is more or > less what we're willing to pay to integrate the (very useful) feature in our > testing setup. that's with -fsanitize=address? as I said, asan allocator uses more memory (redzones, quarantine) and has extra overhead ([un]poisoning redzones, etc) compared to plain lsan. So 20% would be quite expected. Pure lsan should have smaller overhead. We are not actively testing pure lsan on large things because we are already testing all our large things with asan and we don't want yet another build config.
