[ovs-dev] [PATCH net-next v4 08/10] net: openvswitch: fix possible memleak on destroy flow-table

Pravin Shelar pshelar at ovn.org
Tue Oct 29 20:27:31 UTC 2019


On Tue, Oct 29, 2019 at 4:31 AM Tonghao Zhang <xiangxia.m.yue at gmail.com> wrote:
>
> On Tue, Oct 29, 2019 at 3:38 PM Pravin Shelar <pshelar at ovn.org> wrote:
> >
> > On Sun, Oct 27, 2019 at 11:49 PM Tonghao Zhang <xiangxia.m.yue at gmail.com> wrote:
> > >
> > > On Thu, Oct 24, 2019 at 3:14 PM Pravin Shelar <pshelar at ovn.org> wrote:
> > > >
> > > > On Tue, Oct 22, 2019 at 7:35 PM Tonghao Zhang <xiangxia.m.yue at gmail.com> wrote:
> > > > >
> > > > > On Tue, Oct 22, 2019 at 2:58 PM Pravin Shelar <pshelar at ovn.org> wrote:
> > > > > >
> > > > ...
> > > >
> > ...
> > > > >  struct sw_flow *ovs_flow_tbl_dump_next(struct table_instance *ti,
> > > > > @@ -400,10 +458,9 @@ static struct table_instance
> > > > > *table_instance_rehash(struct table_instance *ti,
> > > > >         return new_ti;
> > > > >  }
> > > > >
> > > > > -int ovs_flow_tbl_flush(struct flow_table *flow_table)
> > > > > +int ovs_flow_tbl_flush(struct flow_table *table)
> > > > >  {
> > > > > -       struct table_instance *old_ti, *new_ti;
> > > > > -       struct table_instance *old_ufid_ti, *new_ufid_ti;
> > > > > +       struct table_instance *new_ti, *new_ufid_ti;
> > > > >
> > > > >         new_ti = table_instance_alloc(TBL_MIN_BUCKETS);
> > > > >         if (!new_ti)
> > > > > @@ -412,16 +469,12 @@ int ovs_flow_tbl_flush(struct flow_table *flow_table)
> > > > >         if (!new_ufid_ti)
> > > > >                 goto err_free_ti;
> > > > >
> > > > > -       old_ti = ovsl_dereference(flow_table->ti);
> > > > > -       old_ufid_ti = ovsl_dereference(flow_table->ufid_ti);
> > > > > +       table_instance_destroy(table, true);
> > > > >
> > > > This would destroy running table causing unnecessary flow miss. Lets
> > > > keep current scheme of setting up new table before destroying current
> > > > one.
> > > >
> > > > > -       rcu_assign_pointer(flow_table->ti, new_ti);
> > ....
> > ...
> > >  /* Must be called with OVS mutex held. */
> > >  void ovs_flow_tbl_remove(struct flow_table *table, struct sw_flow *flow)
> > >  {
> > > @@ -752,17 +794,7 @@ void ovs_flow_tbl_remove(struct flow_table
> > > *table, struct sw_flow *flow)
> > >         struct table_instance *ufid_ti = ovsl_dereference(table->ufid_ti);
> > >
> > >         BUG_ON(table->count == 0);
> > > -       hlist_del_rcu(&flow->flow_table.node[ti->node_ver]);
> > > -       table->count--;
> > > -       if (ovs_identifier_is_ufid(&flow->id)) {
> > > -               hlist_del_rcu(&flow->ufid_table.node[ufid_ti->node_ver]);
> > > -               table->ufid_count--;
> > > -       }
> > > -
> > > -       /* RCU delete the mask. 'flow->mask' is not NULLed, as it should be
> > > -        * accessible as long as the RCU read lock is held.
> > > -        */
> > > -       flow_mask_remove(table, flow->mask);
> > > +       table_instance_remove(table, ti, ufid_ti, flow, true);
> > >  }
> > Lets rename table_instance_remove() to imply it is freeing a flow.
> hi Pravin, the function ovs_flow_free will free the flow actually. In
> -ovs_flow_cmd_del
> ovs_flow_tbl_remove
> ...
> ovs_flow_free
>
> In -table_instance_destroy
> table_instance_remove
> ovs_flow_free
>
> But if rename the table_instance_remove, table_instance_flow_free ?
table_instance_flow_free() looks good.


More information about the dev mailing list