Meter bucket configuration has been fixed in ovs commit
7d742b509dd7 ("openvswitch: meter: remove rate from the
bucket size calculation"). Fix CoPP system test failures due to new ovs
datapath behaviour.
Reported-at: https://bugzilla.redhat.com/show_bug.cgi?id=2073060
Signed-off-by: Lorenzo Bianco
cc: ovs-dev (it got dropped by accident)
On 4/12/22 00:07, Terry Wilson wrote:
> On Fri, Apr 1, 2022 at 9:02 AM Dumitru Ceara wrote:
>>
>> On 3/25/22 18:37, Terry Wilson wrote:
>>> Before 46d44cf3b, it was technically possible to assign a monitor
>>> condition directly to Idl.tables[table_name].c
On Thu, Apr 7, 2022 at 4:13 AM Peng He wrote:
>
> I understand there is no need to introduce a rcu api,
> but at least put these codes into a separate function ? :-)
No strong opinion.
>> diff --git a/ofproto/ofproto-dpif.c b/ofproto/ofproto-dpif.c
>> index 6601f2346..402cdcdf5 100644
>> --- a/
On Tue, Apr 12, 2022 at 4:17 AM Peng He wrote:
>
> Hi,
>
> with make check -C build-asan TESTSUITEFLAGS="-k meter -d"
> the ovs is build with asan enabled, is this feature already in current
> building system?
>
> I did not find it. Can anyone give some hint?
> Thanks !
I used a similar configura
David Marchand 于2022年4月12日周二 16:40写道:
> On Thu, Apr 7, 2022 at 4:13 AM Peng He wrote:
> >
> > I understand there is no need to introduce a rcu api,
> > but at least put these codes into a separate function ? :-)
>
> No strong opinion.
>
>
> >> diff --git a/ofproto/ofproto-dpif.c b/ofproto/ofprot
On 4/12/22 00:22, Terry Wilson wrote:
> Before 46d44cf3b, it was technically possible to assign a monitor
> condition directly to Idl.tables[table_name].condition. If done
> before the connection was established, it would successfully apply
> the condition (where cond_change() actually would fail).
Thanks!
David Marchand 于2022年4月12日周二 16:49写道:
> On Tue, Apr 12, 2022 at 4:17 AM Peng He wrote:
> >
> > Hi,
> >
> > with make check -C build-asan TESTSUITEFLAGS="-k meter -d"
> > the ovs is build with asan enabled, is this feature already in current
> > building system?
> >
> > I did not find it
On 4/12/22 10:02, Lorenzo Bianconi wrote:
> Meter bucket configuration has been fixed in ovs commit
> 7d742b509dd7 ("openvswitch: meter: remove rate from the
> bucket size calculation"). Fix CoPP system test failures due to new ovs
> datapath behaviour.
>
> Reported-at: https://bugzilla.redhat.com
On 4/6/22 16:38, Kumar Amber wrote:
> Hashing Optimization are also included which can further
> improve performance by approximately 10%.
Hi, Amber, others. That's an interesting patch set.
I got my hands on an AVX512-capable system for a few hours,
so I was able to play with it. A few observat
Hi Eelco, Mike, all,
Changes in v3
* Added unit test
Changes in v2
* rename parameter "all_slaves_active" to "all_members_active"
* fix doc warnings
Thanks,
Christophe
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/lis
This config param allows the delivery of broadcast and multicast packets
to the secondary interface of non-lacp bonds, equivalent to the option
"all_slaves_active" for kernel bonds.
Reported-at: https://bugzilla.redhat.com/show_bug.cgi?id=1720935
Signed-off-by: Christophe Fontaine
---
NEWS
Hi, folks.
Found some issues while running tests for the hashing optimization
yesterday with UBsan enabled:
lib/dpif-netdev-avx512.c:186:41: runtime error: left shift of 1 by 31 places
cannot be represented in type 'int'
#0 0x20cbf5b in dp_netdev_input_outer_avx512 lib/dpif-netdev-avx512.c:1
The github actions API is currently undergoing some kind of outage - and
has been for the last 12-14 hours (from what I can see). This
particular issue seems to impact how we can retrieve results from
various runs. I am working towards a workaround that will let us move
past this.
GitHub is awar
When compiled with '-fsanitize=undefined' running 'ovsdb-client
--timestamp monitor Open_vSwitch' in a sandbox triggers the following
undefined behavior (flagged by UBSan):
lib/dynamic-string.c:207:38: runtime error: applying zero offset to null
pointer
#0 0x196c05b in ds_put_strftime_mse
The mfex pcap generation script is improved for varied length
traffic and also removes the hard coded mfex_pcap and instead uses
the script itself to generate complex traffic patterns for testing.
Signed-off-by: Kumar Amber
Acked-by: Cian Ferriter
---
tests/automake.mk | 1 -
tests/m
On Tue, Apr 12, 2022 at 3:25 AM Dumitru Ceara wrote:
>
> cc: ovs-dev (it got dropped by accident)
>
> On 4/12/22 00:07, Terry Wilson wrote:
> > On Fri, Apr 1, 2022 at 9:02 AM Dumitru Ceara wrote:
> >>
> >> On 3/25/22 18:37, Terry Wilson wrote:
> >>> Before 46d44cf3b, it was technically possible t
> -Original Message-
> From: dev On Behalf Of Cian Ferriter
> Sent: Thursday, March 3, 2022 10:35 PM
> To: ovs-dev@openvswitch.org
> Subject: [ovs-dev] [PATCH v3 1/6] dpif-netdev-private-extract: Fix typo
> VMBI -> VBMI.
>
> Signed-off-by: Cian Ferriter
Hi Cian,
Thanks for the patch.
For packets which don't already have a hash calculated,
miniflow_hash_5tuple() calculates the hash of a packet
using the previously built miniflow.
This commit adds IPv4 profile specific hashing which
uses fixed offsets into the packet to improve hashing
performance.
Signed-off-by: Harry van Haar
Hi Ilya,
Thanks a lot for such a wonderful review of the series.
Replies inline.
> -Original Message-
> From: Ilya Maximets
> Sent: Tuesday, April 12, 2022 5:05 PM
> To: Amber, Kumar ; ovs-dev@openvswitch.org
> Cc: i.maxim...@ovn.org; echau...@redhat.com; ktray...@redhat.com;
> Ferriter,
No change in behaviour. This is so the MTU to mbuf size
calculations can be done in netdev_dpdk or dpdk code.
Signed-off-by: Kevin Traynor
---
lib/dpdk-stub.c | 8
lib/dpdk.c| 17 +
lib/dpdk.h| 18 ++
lib/netdev-dpdk.c | 32 +---
Shared memory mempools may be currently be shared between DPDK
ports based on port MTU and NUMA. With some hint from the user
we can increase the sharing on MTU and hence reduce memory
consumption in many cases.
For example, a port with MTU 9000, uses a mempool with an
mbuf size based on 9000 MTU.
This patchset optimizes for two cases when using shared mempools.
If there are ports with different MTUs, that usually leads
to multiple shared mempools being created because mempool
mbuf size and hence creation is based from MTU.
In fact, a port with a smaller MTU could share a mempool with
mbuf
> -Original Message-
> From: dev On Behalf Of Cian Ferriter
> Sent: Thursday, March 3, 2022 10:35 PM
> To: ovs-dev@openvswitch.org
> Subject: [ovs-dev] [PATCH v3 2/6] dpif-netdev-lookup: Fix GCC 5 warning.
>
> GCC 5 gave an incompatible pointer type warning for pkt_blocks when it's
> p
Noticed these while doing some other work on docs.
Kevin Traynor (2):
vswitchd.xml: Fix whitespace.
Documentation: Fix use of rst verbatim code chunk syntax.
Documentation/howto/vtep.rst | 1 -
Documentation/internals/contributing/coding-style.rst | 2 --
Documentati
My xml editor keeps autofixing these which means I have to be
careful during 'git add' for unrelated changes. Might as well
just fix them.
Signed-off-by: Kevin Traynor
---
vswitchd/vswitch.xml | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/vswitchd/vswitch.xml b/vswitchd
In some places it is using Markdown syntax and in others
it is not needed as there is already a code block.
Signed-off-by: Kevin Traynor
---
Documentation/howto/vtep.rst | 1 -
Documentation/internals/contributing/coding-style.rst | 2 --
Documentation/topics/dpdk/memory
> -Original Message-
> From: dev On Behalf Of Cian Ferriter
> Sent: Thursday, March 3, 2022 10:35 PM
> To: ovs-dev@openvswitch.org
> Subject: [ovs-dev] [PATCH v3 3/6] dpif-netdev-extract: Remove ISA compiler
> target for mfex.
>
> There are no instructions from this ISA used. The target
Bleep bloop. Greetings Kevin Traynor, I am a robot and I have tried out your
patch.
Thanks for your contribution.
I encountered some error that I wasn't expecting. See the details below.
build:
/bin/sh ./libtool --tag=CC --mode=compile gcc -std=gnu99 -DHAVE_CONFIG_H -I.
-I ./include -I
Bleep bloop. Greetings Kevin Traynor, I am a robot and I have tried out your
patch.
Thanks for your contribution.
I encountered some error that I wasn't expecting. See the details below.
Patch skipped due to previous failure.
Please check this out. If you feel there has been an error, pleas
On Mon, Apr 11, 2022 at 12:37 PM Frode Nordahl
wrote:
>
> tor. 7. apr. 2022, 19:40 skrev Frode Nordahl :
>
> > On Thu, Apr 7, 2022 at 6:51 PM Numan Siddique wrote:
> > >
> > > On Wed, Apr 6, 2022 at 12:11 PM Numan Siddique wrote:
> > > >
> > > > On Wed, Apr 6, 2022 at 4:58 AM Frode Nordahl
> > >
There's no need to call ovn_port_get_peer() to find the peer port of a
logical switch port that's connected to a logical router. We already
store those in op->peer.
Also, factor out addition of router ports to a logical switch's
ovn_datapath and use x2nrealloc().
Signed-off-by: Dumitru Ceara
--
On Tue, Apr 12, 2022 at 6:08 AM Dumitru Ceara wrote:
>
> On 4/12/22 10:02, Lorenzo Bianconi wrote:
> > Meter bucket configuration has been fixed in ovs commit
> > 7d742b509dd7 ("openvswitch: meter: remove rate from the
> > bucket size calculation"). Fix CoPP system test failures due to new ovs
> >
Before 46d44cf3b, it was technically possible to assign a monitor
condition directly to Idl.tables[table_name].condition. If done
before the connection was established, it would successfully apply
the condition (where cond_change() actually would fail).
Although this wasn't meant to be supported,
Hi Ilya,
Sorry for top posting but I wanted to leave a note on my progress.
I have some patches to disable the Linux builds and remove the kernel
module specific specification files for RHEL/CENTOS and the associated
documentation updates. There are also a number of automake.mk files
that I ha
When OVS sees a tunnel push with a nested list next, it will not
clone the packet, as a clone is not needed. However, a clone action will
still be created with the tunnel push encapulated inside. There is no
need to create the clone action in this case, as extra parsing will need
to be performed, w
There's a discussion to have on the intended behavior of a port with 2+
chassis when the switch is attached to localnet.
In general, the patch series makes all chassis switch to tunneling when
multiple chassis are set. (See patch 11/15 "Clone packets to both port
chassis" of the series.) This
From: wenxu
The netdev-offload in tc mode can't work with flow-restore-wait.
When the vswitchd restart with flow-restore-wait, the tc qdisc
will be delete in netdev_set_flow_api_enabled. The netdev flow
api can be enabled after the flow-restore-wait flag removing.
Signed-off-by: wenxu
---
lib/
During incremental flow processing, we track the OVS desired flow
changes so that we can incrementally install them to OVS. The function
merge_tracked_flows() is to merge the "delete and add/update" for the
same flows, to avoid unnecessary changes to OVS when flows are deleted
but added back in the
On Mon, Apr 11, 2022 at 2:05 PM Numan Siddique wrote:
>
> On Sat, Feb 5, 2022 at 12:12 AM Han Zhou wrote:
> >
> > On Fri, Feb 4, 2022 at 7:58 AM Mark Michelson
wrote:
> > >
> > > As far as I can tell, this looks correct.
> > >
> > > Acked-by: Mark Michelson
> >
> > Thanks Mark. I applied it to
39 matches
Mail list logo