Bleep bloop. Greetings Daniel Ding, 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.
checkpatch:
WARNING: The subject summary should end with a dot.
Subject: ovs-tcpdump: Cleanup mirror failed
After running ovs-tcpdump and inputs multiple CTRL+C, the program will
raise the following exception.
Error in atexit._run_exitfuncs:
Traceback (most recent call last):
File "/usr/bin/ovs-tcpdump", line 421, in cleanup_mirror
ovsdb = OVSDB(db_sock)
File "/usr/bin/ovs-tcpdump", line 168,
On Tue, Feb 20, 2024 at 5:35 PM Ilya Maximets wrote:
>
> It's not a system test as it runs with dummy datapath and ports
> and it has nothing to do with layer 3 tunnels.
>
> It should be with other userspace tunnel tests.
>
> While moving also making it a little nicer visually and less error
>
On Tue, Feb 20, 2024 at 8:56 PM Mike Pattrick wrote:
>
> This patch adopts the proposed RFC 6935 by allowing null UDP checksums
> even if the tunnel protocol is IPv6. This is already supported by Linux
> through the udp6zerocsumtx tunnel option. It is disabled by default and
> IPv6 tunnels are
On Tue, Feb 20, 2024 at 10:07 AM Mike Pattrick wrote:
>
> Some network cards support inner checksum offloading but not outer
> checksum offloading. Currently OVS will resolve that outer checksum but
> allows the network card to resolve the inner checksum, invalidating the
> outer checksum in the
When sending packets that are flagged as requiring segmentation to an
interface that doens't support this feature, send the packet to the TSO
software fallback instead of dropping it.
Signed-off-by: Mike Pattrick
---
lib/dp-packet-gso.c | 73 +
This patch adopts the proposed RFC 6935 by allowing null UDP checksums
even if the tunnel protocol is IPv6. This is already supported by Linux
through the udp6zerocsumtx tunnel option. It is disabled by default and
IPv6 tunnels are flagged as requiring a checksum, but this patch enables
the user
On Fri, Feb 16, 2024 at 2:50 PM Numan Siddique wrote:
>
> On Mon, Feb 12, 2024 at 2:49 PM Han Zhou wrote:
> >
> > The MFF_LOG_ENCAP_ID register was defined to save the encap ID and avoid
> > changing across pipelines, but in the function
> > load_logical_ingress_metadata it was reset
On 2/20/24 23:33, Alin Serdean wrote:
> Thank you so much for the patches and dealing with this, Ilya!
>
> Regarding the MD / MT linking for a given binary, this can work, usually,
> without any issues, although it will increase the memory used by the process
> because both runtimes have to be
Local IP is taken into account only in case of IPv4 address, IPv6
source is not checked. That leads to source being ignored during the
route lookup and ultimately packets encapsulated with a source IP
found during a route lookup, which is likely the wrong one.
Even worse, after encapsulation we
From: Ihar Hrachyshka
To mimic what kernel routing subsystem does [1], add a local route
entry for every dummy IP address. This helps with OVN testing multiple
chassis on a single host and allows to run better unit tests for
userspace tunnels without adding route entries manually. This is also
It's not a system test as it runs with dummy datapath and ports
and it has nothing to do with layer 3 tunnels.
It should be with other userspace tunnel tests.
While moving also making it a little nicer visually and less error
prone by requesting port numbers for all the ports.
Signed-off-by:
The whole patch set is needed for a two-line fix in the third patch,
more precisely, for the ability to test these two lines.
First patch in the set is not necessary, but it makes a second patch
a little cleaner.
The second patch is a rebased an taken one step further original patch
from Ihar:
Thank you so much for the patches and dealing with this, Ilya!
Regarding the MD / MT linking for a given binary, this can work, usually,
without any issues, although it will increase the memory used by the process
because both runtimes have to be loaded.
We are linking the rest of the
From: Kevin Sprague
During normal operations, it is useful to understand when a particular flow
gets removed from the system. This can be useful when debugging performance
issues tied to ofproto flow changes, trying to determine deployed traffic
patterns, or while debugging dynamic systems where
All supported versions of Fedora do package libbpf, so it
makes sense to enable USDT support.
Acked-by: Simon Horman
Signed-off-by: Aaron Conole
---
rhel/openvswitch-fedora.spec.in | 8
1 file changed, 8 insertions(+)
diff --git a/rhel/openvswitch-fedora.spec.in
Resurrecting a feature from 2022, introduce a probe that indicates
why a particular flow may be selected for eviction during revalidation
and includes the flow information.
The second patch tells fedora builds to include the USDT probe support
on Fedora systems.
Aaron Conole (1):
rhel: Enable
On 2/20/24 13:43, Eelco Chaudron wrote:
On 19 Feb 2024, at 19:57, Aaron Conole wrote:
Eelco Chaudron writes:
On 12 Feb 2024, at 15:15, Aaron Conole wrote:
Aaron Conole writes:
Eelco Chaudron writes:
On 2 Feb 2024, at 11:31, Adrian Moreno wrote:
On 2/1/24 10:02, Eelco Chaudron
Eelco Chaudron writes:
> On 19 Feb 2024, at 19:57, Aaron Conole wrote:
>
>> Eelco Chaudron writes:
>>
>>> On 12 Feb 2024, at 15:15, Aaron Conole wrote:
>>>
Aaron Conole writes:
> Eelco Chaudron writes:
>
>> On 2 Feb 2024, at 11:31, Adrian Moreno wrote:
>>
>>> On
Jakub Kicinski writes:
> On Fri, 16 Feb 2024 10:28:39 -0500 Aaron Conole wrote:
>> The series is a host of cleanups to the openvswitch selftest suite
>> which should be ready to run under the netdev selftest runners using
>> vng. For now, the testing has been done with RW directories, but
>>
On Tue, Feb 20, 2024 at 4:32 AM Eelco Chaudron wrote:
>
> Avoid unnecessary thread creation as no upcalls are generated,
> resulting in idle threads waiting for process termination.
>
> This optimization significantly reduces memory usage, cutting it
> by half on a 128 CPU/thread system during
On 2/19/24 15:32, Dumitru Ceara wrote:
> On 1/9/24 13:30, Dumitru Ceara wrote:
>> I went ahead and scheduled a new instance of the meeting for:
>>
>> Date/Time: Monday February 19th 16:00 UTC
>> Meeting link: meet.google.com/zns-gqsd-jdn
>> Meeting notes:
>>
On Tue, Feb 20, 2024 at 1:50 AM Adrian Moreno wrote:
>
> In order to properly balance bond traffic, ofproto/bond periodically
> reads usage statistics of the post-recirculation rules (which are added
> to a hidden internal table).
>
> To do that, each "struct bond_entry" (which represents a hash
Some network cards support inner checksum offloading but not outer
checksum offloading. Currently OVS will resolve that outer checksum but
allows the network card to resolve the inner checksum, invalidating the
outer checksum in the process.
Now if we can't offload outer checksums, we don't
On 19 Feb 2024, at 20:23, Ilya Maximets wrote:
> On 2/8/24 16:53, Eelco Chaudron wrote:
>>
>>
>> On 8 Feb 2024, at 15:00, Ilya Maximets wrote:
>>
>>> On 2/8/24 13:44, Eelco Chaudron wrote:
On 6 Feb 2024, at 16:01, Ilya Maximets wrote:
> On 2/6/24 15:07, Eelco Chaudron
On 30 Jan 2024, at 10:53, Simon Horman wrote:
> On Mon, Jan 29, 2024 at 01:51:42PM +0100, Eelco Chaudron wrote:
>> This changes add statistics for when a thread is put into stop state.
>> For example with the following:
>>
>> kill -STOP $(pidof ovs-vswitchd); sleep 1; kill -CONT $(pidof
On 19 Feb 2024, at 19:57, Aaron Conole wrote:
> Eelco Chaudron writes:
>
>> On 12 Feb 2024, at 15:15, Aaron Conole wrote:
>>
>>> Aaron Conole writes:
>>>
Eelco Chaudron writes:
> On 2 Feb 2024, at 11:31, Adrian Moreno wrote:
>
>> On 2/1/24 10:02, Eelco Chaudron wrote:
Avoid unnecessary thread creation as no upcalls are generated,
resulting in idle threads waiting for process termination.
This optimization significantly reduces memory usage, cutting it
by half on a 128 CPU/thread system during testing, with the number
of threads reduced from 95 to 0.
28 matches
Mail list logo