When you set the tx/rx queue number to 10000(or lager), ovs-vswitchd takes more time to process a port with so many rx/tx queues and the ovs_rcu will report the WARN. The 1024 may be appropriate value because the dpdk uses it as default max value. However, 2048 is also ok, just a limiter.
Thanks. Nick > On Jan 9, 2017, at 11:52 PM, Aaron Conole <[email protected]> wrote: > >> This patch avoids the ovs_rcu to report WARN, caused by blocked >> for a long time, when ovs-vswitchd processes a port with many >> rx/tx queues. The number of tx/rx queues per port may be appropriate, >> because the dpdk uses it as an default max value. >> >> Signed-off-by: nickcooper-zhangtonghao <[email protected] >> <mailto:[email protected]>> >> --- > > I don't understand how this change impacts the ovs_rcu warning, unless > you find an issue creating 1025 tx/rx queues. Is that what happened? _______________________________________________ dev mailing list [email protected] https://mail.openvswitch.org/mailman/listinfo/ovs-dev
