On Tue, Jul 05, 2016 at 07:30:11AM -0700, William Tu wrote:
> Hi Ben,
>
> Thanks, this indeed fixes the "possible" leaks.
Thanks, I applied this to master as follows.
--8<--cut here-->8--
From: Ben Pfaff
Date: Tue, 5 Jul 2016 08:33:05 -0700
Subje
Hi Ben,
Thanks, this indeed fixes the "possible" leaks.
Regards,
William
On Fri, Jul 1, 2016 at 8:24 PM, Ben Pfaff wrote:
> Hi William, please try this patch as a substitute for yours. It should
> ensure that pointers to nln_notifiers are to the beginning of the
> structs instead of to the mi
Hi William, please try this patch as a substitute for yours. It should
ensure that pointers to nln_notifiers are to the beginning of the
structs instead of to the middle, meaning that valgrind does not
consider them "possible" leaks.
diff --git a/lib/netlink-notifier.c b/lib/netlink-notifier.c
in
Hi Cascardo,
Thanks for your feedback.
I did a couple of more tests and I think it should be valgrind's false
positive. Even for testcase 1 (TESTSUITEFLAGS='1'), my valgrind
complains about this case as "possible lost."
On the other hand, I do check and make sure that we only called the
name_tabl
On Mon, Jun 20, 2016 at 07:32:52AM -0700, William Tu wrote:
> Testcase 2050, ovn -- 3 HVs, 1 LS, 3 lports/HV, reports possible leak:
> nln_notifier_create (netlink-notifier.c:131)
> name_table_init (route-table.c:319)
> route_table_init (route-table.c:110)
> dp_initialize (dpif.c:12
Testcase 2050, ovn -- 3 HVs, 1 LS, 3 lports/HV, reports possible leak:
nln_notifier_create (netlink-notifier.c:131)
name_table_init (route-table.c:319)
route_table_init (route-table.c:110)
dp_initialize (dpif.c:126)
dp_unregister_provider (dpif.c:218)
dpif_dummy_override (dp
