[ovs-dev] [PATCH v2 00/11] ovn: get rid of most uses of patch ports
Guru Shetty
guru at ovn.org
Mon Dec 19 16:59:00 UTC 2016
On 17 December 2016 at 15:50, Mickey Spiegel <mickeys.dev at gmail.com> wrote:
>
> On Fri, Dec 16, 2016 at 2:51 PM, Guru Shetty <guru at ovn.org> wrote:
>
>> On 16 December 2016 at 14:25, Ben Pfaff <blp at ovn.org> wrote:
>>
>> > At least a v3 will be forthcoming to incorporate Darrell and Liran's
>> > "datapaths of interest" concept, but I said yesterday that I'd send a
>> > revised version before v3 is ready.
>> >
>> > v1->v2:
>> > - Fixed some bugs reported by Mickey throughout the series.
>> > - Patches 1 and 2 are new.
>> > - Patch 3 is modified in various ways, most notably to avoid using
>> the
>> > datapath "sample" action, to add a test, and to better conform to
>> style
>> > in a few small ways.
>> >
>>
>> I was hoping that Mickey's fixes would fix the gateway issues. But I still
>> get gateway test failures when run with:
>> make check-kernel TESTSUITEFLAGS="-k ovn"
>>
>> I will take a closer look, before v3.
>>
>
> I found the problem. The "l3gateway" ports were no longer being added
> to mc groups.
>
> The following incremental fixes the problem:
>
> diff --git a/ovn/controller/physical.c b/ovn/controller/physical.c
> index 8340d54..3c813b9 100644
> --- a/ovn/controller/physical.c
> +++ b/ovn/controller/physical.c
> @@ -630,7 +630,8 @@ consider_mc_group(enum mf_field_id mff_ovn_geneve,
> put_resubmit(OFTABLE_CHECK_LOOPBACK, remote_ofpacts_p);
> } else if (simap_contains(&localvif_to_ofport,
> (port->parent_port && *port->parent_port)
> - ? port->parent_port : port->logical_port)) {
> + ? port->parent_port : port->logical_port)
> + || !strcmp(port->type, "l3gateway")) {
> put_load(port->tunnel_key, MFF_LOG_OUTPORT, 0, 32, ofpacts_p);
> put_resubmit(OFTABLE_CHECK_LOOPBACK, ofpacts_p);
> } else if (port->chassis && !get_localnet_port(local_datapaths,
>
> Mickey
>
Thanks. That looks to be it. The tests pass now.
>
>
>>
>>
>> >
>> > Ben Pfaff (10):
>> > ofp-actions: Use struct ext_action_header for appropriate actions.
>> > ofp-actions: Move function for struct ofpact_nest near struct
>> > definition.
>> > ovn-controller: Make indexes more broadly available.
>> > lport: Be a little more careful building lport index.
>> > lport: Tolerate null pointers in destroy functions.
>> > lport: Add index for logical datapaths.
>> > ovn-controller: Handle only relevant ports and flows.
>> > pnysical: Factor code out of consider_port_binding().
>> > ovn-controller: Avoid code duplication getting chassis record.
>> > ovn-controller: Drop most uses of OVS patch ports.
>> >
>> > William Tu (1):
>> > ofp-actions: Add clone action.
>> >
>> > include/openvswitch/ofp-actions.h | 17 +-
>> > lib/ofp-actions.c | 187 ++++++++++++++-------
>> > ofproto/ofproto-dpif-xlate.c | 14 ++
>> > ovn/controller/binding.c | 118 ++++++++++---
>> > ovn/controller/binding.h | 11 +-
>> > ovn/controller/chassis.c | 12 +-
>> > ovn/controller/chassis.h | 3 +-
>> > ovn/controller/lflow.c | 46 +-----
>> > ovn/controller/lflow.h | 1 -
>> > ovn/controller/lport.c | 82 +++++++++
>> > ovn/controller/lport.h | 33 +++-
>> > ovn/controller/ovn-controller.c | 77 ++++-----
>> > ovn/controller/ovn-controller.h | 33 ++--
>> > ovn/controller/patch.c | 110 ++----------
>> > ovn/controller/patch.h | 6 +-
>> > ovn/controller/physical.c | 340 ++++++++++++++++++++++--------
>> > --------
>> > ovn/controller/physical.h | 9 +-
>> > ovn/controller/pinctrl.c | 36 ++--
>> > ovn/controller/pinctrl.h | 4 +-
>> > tests/ofp-actions.at | 5 +
>> > tests/ofproto-dpif.at | 19 +++
>> > tests/ovn-controller.at | 50 +-----
>> > tests/system-traffic.at | 29 ++++
>> > 23 files changed, 724 insertions(+), 518 deletions(-)
>> >
>> > --
>> > 2.10.2
>> >
>> > _______________________________________________
>> > dev mailing list
>> > dev at openvswitch.org
>> > https://mail.openvswitch.org/mailman/listinfo/ovs-dev
>> >
>> _______________________________________________
>> dev mailing list
>> dev at openvswitch.org
>> https://mail.openvswitch.org/mailman/listinfo/ovs-dev
>>
>
>
More information about the dev
mailing list