On 6/17/20 9:58 AM, wangjian (AN) wrote:
1. Yes, our testers opened some compilation options, such as -fsanitizer=leak.
2. The version of libvirt we are using is 3.2.0. Our tester deployed a long-lived
environment and did many operations (he didn't remember what he did),
such as restarting virtlogd, restarting libvirtd, creating virtual machines, kill -9
virtlogd, libvirtd, etc.
I've tried -fsanitize=leak but the only issues it found were some tests
failing because of unexpected retvals from virsh (because asan changes
that if a memleak was found).
Anyway, I haven't found any leaks inside the virtlogd. The thing is, if
you found memleaks in 3.2.0 that's a shame, but it's likely that those
are fixed upstream already. I mean, 3.2.0 is three years old. Therefore,
we need to prove that these memleaks can happen with more recent
versions and it's not obvious from the code, sorry.
Michal