On 7/17/23 11:13, Ales Musil wrote: > On Tue, Jun 20, 2023 at 3:59 PM Xavier Simonart <[email protected]> wrote: > >> Before this patch, if iface did not exist when qos I+P was run the first >> time >> a qos queue was still created and qos I+P was not applying qos to OVS. >> qos was potentially applied later, if/when (for any reason) runtime_data >> had to be recomputed. >> >> Fixes: 7d1d111ff213 ("controller: configure qos through ovs qos table and >> do not run tc directly") >> Signed-off-by: Xavier Simonart <[email protected]> >> >> --- >> v2: - handled Lorenzo's comments >> - updated test case to also test null network (as referenced by >> Lorenzo) >> - rebased on origin/main >> --- >> controller/binding.c | 26 ++++++++++++-------------- >> tests/ovn.at | 8 ++++++++ >> 2 files changed, 20 insertions(+), 14 deletions(-) >> >> diff --git a/controller/binding.c b/controller/binding.c >> index 8069a2e0d..355b73070 100644 >> --- a/controller/binding.c >> +++ b/controller/binding.c >> @@ -362,7 +362,7 @@ configure_qos(const struct sbrec_port_binding *pb, >> struct qos_queue *q = find_qos_queue(b_ctx_out->qos_map, hash, >> pb->logical_port); >> if (!q || q->min_rate != min_rate || q->max_rate != max_rate || >> - q->burst != burst) { >> + q->burst != burst || (network && strcmp(network, q->network))) { >> struct shash bridge_mappings = >> SHASH_INITIALIZER(&bridge_mappings); >> add_ovs_bridge_mappings(b_ctx_in->ovs_table, >> b_ctx_in->bridge_table, >> &bridge_mappings); >> @@ -378,22 +378,20 @@ configure_qos(const struct sbrec_port_binding *pb, >> add_ovs_qos_table_entry(b_ctx_in->ovs_idl_txn, port, min_rate, >> max_rate, burst, queue_id, >> pb->logical_port); >> + if (!q) { >> + q = xzalloc(sizeof *q); >> + hmap_insert(b_ctx_out->qos_map, &q->node, hash); >> + q->port = xstrdup(pb->logical_port); >> + q->queue_id = queue_id; >> + } >> + free(q->network); >> + q->network = network ? xstrdup(network) : NULL; >> + q->min_rate = min_rate; >> + q->max_rate = max_rate; >> + q->burst = burst; >> } >> shash_destroy(&bridge_mappings); >> } >> - >> - if (!q) { >> - q = xzalloc(sizeof *q); >> - hmap_insert(b_ctx_out->qos_map, &q->node, hash); >> - q->port = xstrdup(pb->logical_port); >> - q->queue_id = queue_id; >> - } >> - >> - free(q->network); >> - q->network = network ? xstrdup(network) : NULL; >> - q->min_rate = min_rate; >> - q->max_rate = max_rate; >> - q->burst = burst; >> } >> >> static const struct ovsrec_queue * >> diff --git a/tests/ovn.at b/tests/ovn.at >> index 2c221a05c..8d69c35d9 100644 >> --- a/tests/ovn.at >> +++ b/tests/ovn.at >> @@ -36235,6 +36235,7 @@ check ovn-nbctl ls-add ls1 >> check ovn-nbctl lsp-add ls1 public1 >> check ovn-nbctl lsp-set-addresses public1 unknown >> check ovn-nbctl lsp-set-type public1 localnet >> +check ovn-nbctl lsp-set-options public1 network_name=phys >> >> check ovn-nbctl lsp-add ls1 lsp5 >> check ovn-nbctl lsp-set-addresses lsp5 f0:00:00:00:00:05 >> @@ -36244,6 +36245,12 @@ ovs-vsctl add-port br-int vif5 -- \ >> ofport-request=5 >> OVS_WAIT_UNTIL([test x`ovn-nbctl lsp-get-up lsp5` = xup]) >> >> +# Delete and add back public1 >> +check ovn-nbctl --wait=hv lsp-del public1 >> +check ovn-nbctl lsp-add ls1 public1 >> +check ovn-nbctl lsp-set-addresses public1 unknown >> +check ovn-nbctl lsp-set-type public1 localnet >> + >> check ovn-nbctl set Logical_Switch_Port public1 >> options:qos_min_rate=6000000000 >> check ovn-nbctl set Logical_Switch_Port public1 >> options:qos_max_rate=7000000000 >> check ovn-nbctl set Logical_Switch_Port public1 >> options:qos_burst=8000000000 >> @@ -36509,6 +36516,7 @@ check ovn-nbctl lsp-add ls2 public2 >> check ovn-nbctl lsp-set-addresses public2 unknown >> check ovn-nbctl lsp-set-type public2 localnet >> check ovn-nbctl --wait=sb set Logical_Switch_Port public2 >> options:qos_min_rate=6000000000 options:qos_max_rate=7000000000 >> options:qos_burst=8000000000 options:network_name=phys >> +check ovn-nbctl --wait=sb lsp-set-options public2 qos_min_rate=6000000000 >> qos_max_rate=7000000000 qos_burst=8000000000 >> >> # Let's now send ovn controller to sleep, so it will receive both ofport >> notification and ls deletion simultaneously >> sleep_controller hv-1 >> -- >> 2.31.1 >> >> _______________________________________________ >> dev mailing list >> [email protected] >> https://mail.openvswitch.org/mailman/listinfo/ovs-dev >> >> > Looks good to me, thanks. > > Acked-by: Ales Musil <[email protected]> >
Thanks, Xavier and Ales! I applied this to main and backported it to 23.06. Regards, Dumitru _______________________________________________ dev mailing list [email protected] https://mail.openvswitch.org/mailman/listinfo/ovs-dev
