>On 2/28/23 17:14, Ilya Maximets wrote:
>> On 2/28/23 17:05, Maxime Coquelin wrote:
>>>
>>>
>>> On 2/28/23 16:37, Ilya Maximets wrote:
On 2/28/23 16:16, Maxime Coquelin wrote:
> Hi,
>
> On 2/28/23 10:33, Wan Junjie wrote:
>> Hi,
>>
>> On Tue, Feb 28, 2023 at 4:03 PM D
On 14/03/2023 13:04, Simon Horman wrote:
> On Mon, Mar 13, 2023 at 07:47:14PM +0200, Roi Dayan wrote:
>>
>>
>> On 13/03/2023 14:16, Simon Horman wrote:
>>> On Mon, Mar 13, 2023 at 12:31:49PM +0200, Roi Dayan wrote:
On 13/03/2023 11:01, Eelco Chaudron wrote:
>
>
> On 13
Assume the following setup:
++
| Logical Router |
| lr001 +-+
++ |
|
++ |
| Logical Router | | ++ +--+
| lr002 +-+-+ Logical Switch +-+ Phyiscal Network |
++ | | ls-
Bleep bloop. Greetings Felix Hüttner, 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:
ERROR: Author Felix Hüttner needs to sign off.
WARNING: Unexpected sign-offs from developers
In commit [1] there was added support to enable memory trimming on OVSDB tlog
compation. However there was no option to get current setting value except
log parsing. This patch adds new unixctl command
'ovsdb-server/show-memory-trim-on-compaction' to print current setting value.
1:
https://gith
On 20 Mar 2023, at 6:44, Chris Mi wrote:
> On 3/16/2023 5:09 PM, Eelco Chaudron wrote:
>> On 1 Mar 2023, at 8:22, Chris Mi wrote:
>>
>>> When offloading sample action to TC, userspace creates a unique ID
>>> to map sFlow action and tunnel info and passes this ID to kernel
>>> instead of the sFlo
This patch set covers removal of expressions which are subsets of other
wider expressions and aggregation of a few granular expressions into
wider expressions that cover all of them at once. This allows to avoid
flow explosion in case of negative matches and reduce the total number
of flows requir
While crushing OR expressions, OVN removes exact replicas of sub
expressions. However, there could be many CMP expressions that are
supersets of each other. These are most likely to be created as a
result of cross-product while expanding brackets in the AND expression
in crush_and_numeric(), i.e.
expr_sort() can potentially return NULL if all sub-expressions got
crushed into 'true'. This didn't happen before, but can happen
with more aggressive optimizations in crush_or().
Signed-off-by: Ilya Maximets
---
lib/expr.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib
If two simple CMP expressions within OR have exact same masks, and
their values differ in a single bit, these expressions can be combined
into one by removing that bit from the mask.
E.g. Expression (tcp.src == 6441 || tcp.src == 6443) can be
simplified down to single tcp.src == 0x1929/0xfffd
On Mon, Mar 20, 2023 at 09:51:48AM +0200, Roi Dayan wrote:
>
>
> On 14/03/2023 13:04, Simon Horman wrote:
> > On Mon, Mar 13, 2023 at 07:47:14PM +0200, Roi Dayan wrote:
> >>
> >>
> >> On 13/03/2023 14:16, Simon Horman wrote:
> >>> On Mon, Mar 13, 2023 at 12:31:49PM +0200, Roi Dayan wrote:
>
On 3/20/23 11:44, Simon Horman wrote:
> On Mon, Mar 20, 2023 at 09:51:48AM +0200, Roi Dayan wrote:
>>
>>
>> On 14/03/2023 13:04, Simon Horman wrote:
>>> On Mon, Mar 13, 2023 at 07:47:14PM +0200, Roi Dayan wrote:
On 13/03/2023 14:16, Simon Horman wrote:
> On Mon, Mar 13, 2023 at 1
The CT flush was enabled by default for every LB, add
config option called "ct_flush" that allows
users to enable/disable the CT flush. The CT flush option
is set to "false" by default.
Reported-at: https://bugzilla.redhat.com/2178962
Acked-by: Numan Siddique
Signed-off-by: Ales Musil
---
v2: Ma
Bleep bloop. Greetings Ilya Maximets, 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: Line is 93 characters long (recommended limit is 79)
#1300 FILE: p-set-addresses:1752
On 20 Mar 2023, at 11:48, Ilya Maximets wrote:
> On 3/20/23 11:44, Simon Horman wrote:
>> On Mon, Mar 20, 2023 at 09:51:48AM +0200, Roi Dayan wrote:
>>>
>>>
>>> On 14/03/2023 13:04, Simon Horman wrote:
On Mon, Mar 13, 2023 at 07:47:14PM +0200, Roi Dayan wrote:
>
>
> On 13/03/202
parse_tc_flower_to_actions() was not reporting errors, which would
cause parse_tc_flower_to_match() to ignore them.
Fixes: dd03672f7bbb ("netdev-offload-tc: Move flower_to_match action handling
to isolated function.")
Signed-off-by: Eelco Chaudron
---
lib/netdev-offload-tc.c | 36
On 3/20/23 11:53, Ales Musil wrote:
> The CT flush was enabled by default for every LB, add
> config option called "ct_flush" that allows
> users to enable/disable the CT flush. The CT flush option
> is set to "false" by default.
>
> Reported-at: https://bugzilla.redhat.com/2178962
> Acked-by: Num
On 3/20/23 10:24, Vladislav Odintsov wrote:
> In commit [1] there was added support to enable memory trimming on OVSDB tlog
> compation. However there was no option to get current setting value except
> log parsing. This patch adds new unixctl command
> 'ovsdb-server/show-memory-trim-on-compactio
Hi Ilya,
Ah, I didn’t see that its default was changed in newer versions...
We run 2.17 where it’s off by default and I decided to submit a patch which is
useful for that version.
Anyway my opinion is if user has an ability to change the setting, it should be
possible to get its actual value. At
On 3/20/23 13:20, Vladislav Odintsov wrote:
> Hi Ilya,
>
> Ah, I didn’t see that its default was changed in newer versions...
> We run 2.17 where it’s off by default and I decided to submit a patch which
> is useful for that version.
> Anyway my opinion is if user has an ability to change the set
No, my usecase with 2.17 is that I just want to check agains a running process
wether memory compaction is enabled or not without searching specific line in
logs, which in addition can be rotated.
> On 20 Mar 2023, at 15:27, Ilya Maximets wrote:
>
> On 3/20/23 13:20, Vladislav Odintsov wrote:
We are performing a regular backup & update on the 0daybot CI host
system today. This means 0daybot, Github Actions and other CI tools will be
down for today
We expect that it will be completed by EOB
Best,
___
dev mailing list
d...@openvswitch.org
http
Hi Ales,
I agree with Ilya about updating the NEWS wording. I also think this is
something that a committer can take care of when committing the change.
Acked-by: Mark Michelson
On 3/20/23 08:00, Ilya Maximets wrote:
On 3/20/23 11:53, Ales Musil wrote:
The CT flush was enabled by default f
Thanks Ales for the patch and Ilya, Mark and Numan for reviews!
I applied it to main with the following NEWS entry rephrase:
diff --git a/NEWS b/NEWS
index 1298d16a23..7edae22ff5 100644
--- a/NEWS
+++ b/NEWS
@@ -2,10 +2,9 @@ Post v23.03.0
-
- Enhance LSP.options:arp_proxy to suppo
The CT flush was enabled by default for every LB, add
config option called "ct_flush" that allows
users to enable/disable the CT flush. The CT flush option
is set to "false" by default.
Reported-at: https://bugzilla.redhat.com/2178962
Acked-by: Numan Siddique
Acked-by: Mark Michelson
Signed-off-
On 3/20/23 17:08, Ales Musil wrote:
> The CT flush was enabled by default for every LB, add
> config option called "ct_flush" that allows
> users to enable/disable the CT flush. The CT flush option
> is set to "false" by default.
>
> Reported-at: https://bugzilla.redhat.com/2178962
> Acked-by: Num
Drop ip packets with ct status set to invalid in post snat and
lb_aff_learn router stages.
Skip ICMPv{4,6} error messages packet in ct.inv rules in order to avoid
to introduce too complicated code.
Reported-at: https://bugzilla.redhat.com/show_bug.cgi?id=2160685
Reviewed-by: Simon Horman
Signed-o
Hi Lorenzo,
Since this is an RFC series, I did not do an in-depth review of the
patches. I just took a quick look.
My only concern is that patch 1 breaks existing configurations for QoS,
since the ovs-port parameter is now required. Is there any way to
preserve existing behavior [1] for depl
On Mon, Mar 20, 2023 at 3:37 AM Ilya Maximets wrote:
>
> This patch set covers removal of expressions which are subsets of other
> wider expressions and aggregation of a few granular expressions into
> wider expressions that cover all of them at once. This allows to avoid
> flow explosion in case
Fix non-tcp haripin use-case if the load-balancer is configured without
ports.
Fixes: 022ea339c8e2 ("lflow: Use learn() action to generate LB hairpin reply
flows.")
Tested-by: Ying Xu
Reviewed-by: Simon Horman
Reported-at: https://bugzilla.redhat.com/show_bug.cgi?id=2157846
Signed-off-by: Loren
The refactor doesn't have any functional changes to the code, but to maintain
function dependencies more clearly at a finer grain level. As a byproduct it
also improves performance slightly.
The refactor touches a big part of northd code, so submit it before the real
incremental processing changes
We already have a helper function for the logic, so reuse it.
Signed-off-by: Han Zhou
---
northd/northd.c | 7 ++-
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/northd/northd.c b/northd/northd.c
index 1e49301dd5b4..8f90e0ce4e04 100644
--- a/northd/northd.c
+++ b/northd/northd
The hmap 'ports' includes both logical switch ports and logical router
ports. Although most of the code handles these two different types of port
differently, we always have to iterate through the whole hmap, and skip
one type of ports in the loop. More importantly, this hides dependency
from the i
Similar to the previous commit that splits switch ports and router
ports, this patch splits the two types of datapaths in different hmap
tables. This is also mainly to help for future incremental processing
implementation. As a byproduct, the recompute performance is improved
slightly. (Together w
The I-P engine node functions get inputs from the node's input data, and
then call the real implementation functions (implemented in northd.c)
with a xxx_input wrapper variable, and then in the called functions, it
passes the variable further down to other functions, regardless of what
fields are r
In the current code, many functions depends on the global datapaths
arrays variables, which is convenient. However, for incremental
processing framework to track dependencies properly, we should avoid
using global variables and purely depends on engine inputs (and function
inputs, for the functions
With the help of the previous refactor that exposes inputs explicitly,
we could tell that there are many unused dependencies in the current I-P
engine nodes. This patch removes the unused ones.
Signed-off-by: Han Zhou
---
northd/en-northd.c | 2 --
northd/inc-proc-northd.c | 62 ++
37 matches
Mail list logo
