> Hi,
> I just upgraded from OvS 2.7 + DPDK 16.11 to OvS2.9 + DPDK 17.11 and
> running into performance issue with 64 Byte packet rate. One interesting
> thing that I notice that even at very light load from IXIA the processing
> cycles on all the PMD threads run close to 100% of the cpu cycle on 2
> On Jun 14, 2018, at 10:40 PM, Darrell Ball wrote:
>
>> On Thu, Jun 14, 2018 at 12:00 PM, Justin Pettit wrote:
>> Signed-off-by: Justin Pettit
>> ---
>> lib/dpctl.c | 158
>> +++-
>> 1 file changed, 38 insertions(+), 120 deletions(-)
>
> On Jun 14, 2018, at 9:41 PM, Ben Pfaff wrote:
>
> On Thu, Jun 14, 2018 at 12:00:43PM -0700, Justin Pettit wrote:
>> Signed-off-by: Justin Pettit
>
> For the series:
> Acked-by: Ben Pfaff
Thanks. I pushed the series to master. I also cherry-picked the first patch
to branch-2.8 and branc
Signed-off-by: Justin Pettit
---
lib/ofp-meter.c | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/lib/ofp-meter.c b/lib/ofp-meter.c
index 4f77f508d396..de6f2b3a0fc2 100644
--- a/lib/ofp-meter.c
+++ b/lib/ofp-meter.c
@@ -603,12 +603,13 @@ parse_ofp_meter_mod_str__(struct
Add TCP reset/ICMP port unreachable messages in reply to IP packets directed to
the logical router's IP addresses
Lorenzo Bianconi (3):
OVN: add UDP port unreachable support to OVN logical router
OVN: add TCP port unreachable support to OVN logical router
OVN: add protocol unreachable suppor
Add priority-80 flows to generate ICMP port unreachable messages in
reply to UDP datagrams directed to the router's IP address since the
logical router doesn't accept any UDP traffic
Signed-off-by: Lorenzo Bianconi
---
ovn/northd/ovn-northd.8.xml | 4 --
ovn/northd/ovn-northd.c | 20 +++
Add priority-80 flows to generate TCP reset messages in reply to
TCP datagrams directed to the router's IP address since the
logical router doesn't accept any TCP traffic
Signed-off-by: Lorenzo Bianconi
---
ovn/northd/ovn-northd.8.xml | 4
ovn/northd/ovn-northd.c | 33 +
Add priority-70 flows to generate ICMP protocol unreachable messages
in reply to packets directed to the router's IP address on IP protocols
other than UDP, TCP, and ICMP
Signed-off-by: Lorenzo Bianconi
---
ovn/northd/ovn-northd.8.xml | 4
ovn/northd/ovn-northd.c | 17 +
Hi Ben, Justin, William and Darrell,
Please see below email regarding performance optimization, I believe we can
speed up the mf_get_next_in_map() and related dpcls_lookup() code significantly
if we can confirm that the suggestion below is valid.
Your input and feedback would be very helpful.
Re
lib/netdev-dpdk.c: In function :
lib/netdev-dpdk.c:2865:49: warning: output may be truncated before the last
format character [-Wformat-truncation=]
snprintf(vhost_vring, 16, "vring_%d_size", i);
^
Since vring_num is 16 bits, the larges
Acked-by: Mark Michelson
On 06/14/2018 03:36 PM, Ben Pfaff wrote:
This makes the comment more meaningful.
Signed-off-by: Ben Pfaff
---
ovn/controller/binding.c | 13 -
1 file changed, 8 insertions(+), 5 deletions(-)
diff --git a/ovn/controller/binding.c b/ovn/controller/bindin
> On Thu, Jun 14, 2018 at 05:27:18PM +0200, Lorenzo Bianconi wrote:
>> Using icmp4 action, send an ICMP time exceeded frame whenever
>> an OVN logical router receives an IPv4 packets whose TTL has
>> expired (ip.ttl == {0, 1})
>>
>> Signed-off-by: Lorenzo Bianconi
>> ---
>> Changes since v1:
>> -
On 06/13/2018 11:29 PM, Han Zhou wrote:
On Wed, Jun 13, 2018 at 3:37 PM, Ben Pfaff wrote:
To make ovn-controller recompute incrementally, we need accurate
dependencies for each function that reads or writes a table. It's
currently difficult to be sure about these dependencies, and certainly
d
> lib/netdev-dpdk.c: In function :
> lib/netdev-dpdk.c:2865:49: warning: output may be truncated before the last
> format character [-Wformat-truncation=]
> snprintf(vhost_vring, 16, "vring_%d_size", i);
> ^
>
> Since vring_num is 16 bi
On Wed, Jun 06, 2018 at 03:17:59PM +0100, Kyle Simpson wrote:
> The previous implementation scans the action set of each WRITE_ACTIONS
> command 13--17 times when moving the actions over. This change builds
> up the list as a single scan, which should be more efficient.
>
> Signed-off-by: Kyle Sim
Thanks, applied to master.
On Fri, Jun 15, 2018 at 09:20:49AM -0400, Mark Michelson wrote:
> Acked-by: Mark Michelson
>
> On 06/14/2018 03:36 PM, Ben Pfaff wrote:
> >This makes the comment more meaningful.
> >
> >Signed-off-by: Ben Pfaff
> >---
> > ovn/controller/binding.c | 13 -
>
On Fri, Jun 15, 2018 at 01:44:50AM -0700, Justin Pettit wrote:
> Signed-off-by: Justin Pettit
Acked-by: Ben Pfaff
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
[Dropping invalid email address findtheonly...@example.com.]
Thanks.
I believe that we've found and fixed related memory errors before. What
version of OVS are you using?
On Fri, Jun 15, 2018 at 02:48:11PM +0800, 孙文杰 wrote:
> Yes, I can.
>
> When the first packets send to sdn contoller(we use
On Fri, Jun 15, 2018 at 05:56:28PM +0300, Ilya Maximets wrote:
> > lib/netdev-dpdk.c: In function :
> > lib/netdev-dpdk.c:2865:49: warning: output may be truncated before the
> > last format character [-Wformat-truncation=]
> > snprintf(vhost_vring, 16, "vring_%d_size", i);
> > ^~
Hi Shahaf,
Thanks for pointing NIC/protocol support.
Find inline comments:
>>When you say DECAP functionality is broken you mean the flow is not actually
>>inserted to the HW right?
[Mallesh]: Yes, DECAP flows are not inserted to HW.
>>The datapath should still DECAP the flow correctly.
[Mal
On Thu, Jun 07, 2018 at 02:59:46PM +0530, vkomm...@redhat.com wrote:
> From: Venkata Anil
>
> This patch avoids tunneling and instead uses source tenant vlan network
> across hypervisors for traffic from vlan network on local hypervisor
> towards gateway hypervisor hosting redirect chassiss port.
On Fri, Jun 15, 2018 at 11:49:06AM +0200, Lorenzo Bianconi wrote:
> Add TCP reset/ICMP port unreachable messages in reply to IP packets directed
> to
> the logical router's IP addresses
Thanks for implementing this!
I noticed a needlessly inefficient pattern here, where the actions are
a fixed s
On Thu, May 31, 2018 at 11:13:19AM +0200, Eelco Chaudron wrote:
> This change adds the ovs_dump_dp_provider command, which allows
> dumping of all the registered registered_dpif_class structures.
>
> In addition it has some small internal cleanups.
>
> Signed-off-by: Eelco Chaudron
Thanks, appl
On Fri, Jun 01, 2018 at 01:21:31PM +0200, Eelco Chaudron wrote:
> This changes add two additional gdb commands:
>
> - ovs_dump_netdev_provider
> - ovs_dump_ovs_list
Thanks, applied to master.
___
dev mailing list
d...@openvswitch.org
https://mail.openv
I think I understand Yipeng's point 2. Please allow me to restate it:
if a flow's field has a 1-bit in its mask, then the field must exist in
the packet for the flow to match. This is true. (There are some
technical exceptions: for example, OVS and OpenFlow handle VLANs in an
unusual way that al
On Wed, May 30, 2018 at 10:08:26AM -0700, Han Zhou wrote:
> In ovsdb_idl_db_track_clear(), it needs to free the deleted row.
> However, it unnecessary to call ovsdb_idl_row_clear_old(), because
> this has been called in ovsdb_idl_row_destroy(). It is also confusing
> because it is called only if:
>
On Tue, May 29, 2018 at 06:06:19PM +, Neal Shrader via dev wrote:
> While investigating a kernel panic, our team noticed that UDP traffic
> recieved by an STT tunnel will always have a gso_type set as SKB_GSO_UDP.
> After decap, we also noticed that traffic that had this flag set had its
> frag
On 6/15/2018 1:05 PM, Ben Pfaff wrote:
On Tue, May 29, 2018 at 06:06:19PM +, Neal Shrader via dev wrote:
While investigating a kernel panic, our team noticed that UDP traffic
recieved by an STT tunnel will always have a gso_type set as SKB_GSO_UDP.
After decap, we also noticed that traffic t
On Fri, Jun 15, 2018 at 1:03 PM, Ben Pfaff wrote:
>
> On Wed, May 30, 2018 at 10:08:26AM -0700, Han Zhou wrote:
> > In ovsdb_idl_db_track_clear(), it needs to free the deleted row.
> > However, it unnecessary to call ovsdb_idl_row_clear_old(), because
> > this has been called in ovsdb_idl_row_dest
Hi, Ben,
Your understand of my point 2 is correct. But more specifically, by "0" or "1"
I mean the bit in "flowmap map" of the miniflow. Just as a remind, here is the
code comment of the flowmap:
* The map member hold one bit for each uint64_t in a "struct flow". Each
* 0-bit indicates that t
When ovsdb-server closes a remote connection, it logs a message about it
that includes the reason. Until now this has included sessions that it
closes when it exits. That meant that, when --run was used, there was a
race between noticing that the subprocess exited and noticing that the
session th
A user reported a failure of test 2364 "vlog - RFC5424 facility - Python2"
with an exit code that says that the test-unixctl process died from an
uncaught exception. Unfortunately the exception didn't show up in the log.
This commit should make the exception show up (it deletes some boilerplate
we
On Fri, Jun 15, 2018 at 1:21 PM, Gregory Rose wrote:
> On 6/15/2018 1:05 PM, Ben Pfaff wrote:
>>
>> On Tue, May 29, 2018 at 06:06:19PM +, Neal Shrader via dev wrote:
>>>
>>> While investigating a kernel panic, our team noticed that UDP traffic
>>> recieved by an STT tunnel will always have a g
Every iteration of the main loop allocated a new IPAM structure.
Found by inspection.
Signed-off-by: Ben Pfaff
---
ovn/northd/ovn-northd.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/ovn/northd/ovn-northd.c b/ovn/northd/ovn-northd.c
index 74eefc6caeba..344b595ad3c1 10
While reviewing https://patchwork.ozlabs.org/patch/924319/, I discovered
some bugs in IPAM that seem worth fixing. The first two patches below are
minimal so that they can be backported. The third is an improvement that
doesn't need backporting.
Ben Pfaff (3):
ovn-northd: Fix memory leak when
The code for doing IPAM never gave up if IPAM was un-configured later.
Found by inspection.
Signed-off-by: Ben Pfaff
---
ovn/northd/ovn-northd.c | 14 ++
1 file changed, 14 insertions(+)
diff --git a/ovn/northd/ovn-northd.c b/ovn/northd/ovn-northd.c
index 344b595ad3c1..8755fa1f40c0
Until now, the ipam_info struct for a datapath has been allocated on
demand. This leads to slightly complication in the code in places, and
there is hardly any benefit since ipam_info is only about 48 bytes anyway.
This commit just inlines it into struct ovn_datapath.
Signed-off-by: Ben Pfaff
--
On Fri, Jun 01, 2018 at 03:41:14PM -0400, Mark Michelson wrote:
> OVN offers a method of IP address management that allows for an IPv4 subnet or
> IPv6 prefix to be specified on a logical switch. Then by specifying a
> switch port's address as "dynamic" or " dynamic", OVN will
> automatically assig
On Tue, May 29, 2018 at 09:15:52AM -0400, Mark Michelson wrote:
> Hi Ben,
>
> Thanks for having a look and providing an update. See my replies in-line
> below.
>
> On 05/25/2018 06:31 PM, Ben Pfaff wrote:
> >On Thu, May 17, 2018 at 01:16:55PM -0400, Mark Michelson wrote:
> >>When inserting data i
ofpacts_check__() was a huge switch statement with special cases for many
different kinds of actions. This made it unwieldy and put the special
cases far away from the rest of the code related to a given action. This
commit refactors the code to avoid the problem.
Signed-off-by: Ben Pfaff
---
I
On Wed, May 23, 2018 at 09:28:59PM -0700, Ben Pfaff wrote:
> On Wed, May 23, 2018 at 06:06:44PM -0700, Han Zhou wrote:
> > On Wed, May 23, 2018 at 5:14 PM, Ben Pfaff wrote:
> > >
> > > Until now, rconn_get_version() has only reported the OpenFlow version in
> > > use when the rconn is actually con
Signed-off-by: Ben Pfaff
---
v1->v2: Rebase due to changes on master.
include/openvswitch/ofp-bundle.h | 11 +-
include/openvswitch/ofp-connection.h | 18 +-
include/openvswitch/ofp-ipfix.h | 5 +
include/openvswitch/ofp-match.h | 9 +-
include/openvswitch/ofp-meter.h |
> I like how it actually deletes more lines (161)
> than it adds (106).
A good result. :)
> What do you think of this version?
Looks good to me, I'm not a macro person so it was strange at first glance.
It makes a lot of sense though, and it's certainly more extensible
than my mechanism and what
On Fri, Jun 15, 2018 at 7:11 AM, Mark Michelson wrote:
> On 06/13/2018 11:29 PM, Han Zhou wrote:
>
>> On Wed, Jun 13, 2018 at 3:37 PM, Ben Pfaff wrote:
>>
>>>
>>> To make ovn-controller recompute incrementally, we need accurate
>>> dependencies for each function that reads or writes a table. It
On Fri, Jun 15, 2018 at 07:22:38PM -0700, Han Zhou wrote:
> On Fri, Jun 15, 2018 at 7:11 AM, Mark Michelson wrote:
>
> > On 06/13/2018 11:29 PM, Han Zhou wrote:
> >
> >> On Wed, Jun 13, 2018 at 3:37 PM, Ben Pfaff wrote:
> >>
> >>>
> >>> To make ovn-controller recompute incrementally, we need acc
On Fri, Jun 15, 2018 at 10:11:41AM -0400, Mark Michelson wrote:
> On 06/13/2018 11:29 PM, Han Zhou wrote:
> >On Wed, Jun 13, 2018 at 3:37 PM, Ben Pfaff wrote:
> >>
> >>To make ovn-controller recompute incrementally, we need accurate
> >>dependencies for each function that reads or writes a table.
46 matches
Mail list logo
