On Tue, Feb 6, 2024 at 6:49 PM Xavier Simonart wrote:
> Hi Ales
>
> Thanks for the review.
> I'll add NEXT(x, y), with x being ingress or egress, and y being lflow
> table number, and use them where necessary.
> I'll post a v2 when the 26 first patches from the series get merged so
> that ovs
Bleep bloop. Greetings Nobuhiro MIKI, 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 82 characters long (recommended limit is 79)
#22 FILE:
Signed-off-by: Nobuhiro MIKI
---
Documentation/tutorials/ovn-sandbox.rst | 18 +-
1 file changed, 9 insertions(+), 9 deletions(-)
diff --git a/Documentation/tutorials/ovn-sandbox.rst
b/Documentation/tutorials/ovn-sandbox.rst
index 2b574c02f153..decc8abb397a 100644
---
On Tue, Feb 6, 2024 at 3:52 PM Han Zhou wrote:
>
> On Mon, Feb 5, 2024 at 7:47 PM Numan Siddique wrote:
> >
> > On Mon, Feb 5, 2024 at 9:41 PM Han Zhou wrote:
> > >
> > > On Mon, Feb 5, 2024 at 4:12 PM Numan Siddique wrote:
> > > >
> > > > On Mon, Feb 5, 2024 at 5:54 PM Han Zhou wrote:
> > >
On 2/6/24 22:22, Ilya Maximets wrote:
>>>
>>> I did some testing with my patch and these are the findings
>>>
>>> | Avg. Poll Intervals | Total
>>> test time | northd RSS
>>> -- ++---
>>> Last
>>
>> I did some testing with my patch and these are the findings
>>
>> | Avg. Poll Intervals | Total
>> test time | northd RSS
>> -- ++---
>> Last week | 1.5 seconds
On Mon, Feb 5, 2024 at 7:47 PM Numan Siddique wrote:
>
> On Mon, Feb 5, 2024 at 9:41 PM Han Zhou wrote:
> >
> > On Mon, Feb 5, 2024 at 4:12 PM Numan Siddique wrote:
> > >
> > > On Mon, Feb 5, 2024 at 5:54 PM Han Zhou wrote:
> > > >
> > > > On Mon, Feb 5, 2024 at 10:15 AM Ilya Maximets
> >
Hi Ales
Thanks for the review.
I'll add NEXT(x, y), with x being ingress or egress, and y being lflow
table number, and use them where necessary.
I'll post a v2 when the 26 first patches from the series get merged so that
ovs robot can run happily.
With this and your 26 patches, we should be
On Tue, Feb 6, 2024 at 3:45 PM Xavier Simonart wrote:
> This patch uses the recently introduced macros defining openflow table
> numbers.
>
> Signed-off-by: Xavier Simonart
> ---
>
Hi Xavier,
thank you for the follow up to remove the rest of hardcoded table numbers.
There are a few places
This patch fixes the correctness of dp_packet_inner_l4_size() when
checking for the existence of an inner L4 header. Previously it checked
for the outer L4 header.
This function is currently only used when a packet is already flagged
for tunneling, so an incorrect determination isn't possible as
Previously the BFD packet creation code did not appropriately set
offsets or flags. This contributed to issues involving encapsulation and
the TSO code.
The transition to using standard functions also means some other
metadata like packet_type are set appropriately.
Fixes: ccc096898c46 ("bfd:
Currently the monitor will call dp_packet_clear() on the dp_packet that
is shared amongst BFD, LLDP, and CFM. However, all of these packets are
created with eth_compose(), which already calls dp_packet_clear().
Reviewed-by: David Marchand
Signed-off-by: Mike Pattrick
---
Include inner offsets in functions where l3 and l4 offsets are either
modified or checked.
Fixes: 084c8087292c ("userspace: Support VXLAN and GENEVE TSO.")
Signed-off-by: Mike Pattrick
---
v2:
- Prints out new offsets in autovalidator
- Extends resize_l2 change to avx512
v3:
- Reordered
Eric Dumazet writes:
> On Tue, Feb 6, 2024 at 3:55 PM Aaron Conole wrote:
>>
>>
>> Oops - I didn't consider it.
>>
>> Given that, maybe the best approach would not to rely on per-cpu
>> counter. I'll respin in the next series with a depth counter that I pass
>> to the function instead and
On Tue, Feb 6, 2024 at 3:55 PM Xavier Simonart wrote:
> Hi Ales
>
Hi Xavier,
>
> Thanks for the series and for making it "review friendly".
> I only had two comments for patches 1 to 26
> - There are a few remaining table numbers in comments which will become
> wrong if table numbers are
Bleep bloop. Greetings Xavier Simonart, 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.
git-am:
error: sha1 information is lacking or useless (tests/ovn.at).
error: could not build fake
On 2/6/24 15:07, Eelco Chaudron wrote:
>
>
> On 6 Feb 2024, at 14:46, Ilya Maximets wrote:
>
>> On 2/6/24 14:30, Eelco Chaudron wrote:
>>> Previously, the ability to override the system default for the number
>>> of handler threads was broken since to the introduction of the per-CPU
>>> upcall
On 6 Feb 2024, at 15:50, David Marchand wrote:
> On Tue, Feb 6, 2024 at 3:47 PM Eelco Chaudron wrote:
>> On 6 Feb 2024, at 15:17, David Marchand wrote:
>>
>>> On Tue, Feb 6, 2024 at 2:31 PM Eelco Chaudron wrote:
Avoid unnecessary thread creation as no upcalls are generated,
On Tue, Feb 6, 2024 at 3:55 PM Aaron Conole wrote:
>
>
> Oops - I didn't consider it.
>
> Given that, maybe the best approach would not to rely on per-cpu
> counter. I'll respin in the next series with a depth counter that I pass
> to the function instead and compare that. I guess that should
Hi Ales
Thanks for the series and for making it "review friendly".
I only had two comments for patches 1 to 26
- There are a few remaining table numbers in comments which will become
wrong if table numbers are changing. Some are oflow tables for lflows, so
should we just remove them / replace
Eric Dumazet writes:
> On Tue, Feb 6, 2024 at 2:11 PM Aaron Conole wrote:
>>
>> The ovs module allows for some actions to recursively contain an action
>> list for complex scenarios, such as sampling, checking lengths, etc.
>> When these actions are copied into the internal flow table, they are
On Tue, Feb 6, 2024 at 3:47 PM Eelco Chaudron wrote:
> On 6 Feb 2024, at 15:17, David Marchand wrote:
>
> > On Tue, Feb 6, 2024 at 2:31 PM Eelco Chaudron wrote:
> >>
> >> Avoid unnecessary thread creation as no upcalls are generated,
> >> resulting in idle threads waiting for process
On 6 Feb 2024, at 15:17, David Marchand wrote:
> On Tue, Feb 6, 2024 at 2:31 PM 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,
There are only two changes:
- The AT_DATA content is not double quoted between square brackets.
- All '[' have been replaced by '[[' and all ']' by ']]'
This patch will be used in subsequent patch to remove hard-code table numbers
Signed-off-by: Xavier Simonart
---
tests/ovn.at | 590
This patch uses the recently introduced macros defining openflow table numbers.
Signed-off-by: Xavier Simonart
---
tests/ovn-macros.at | 4 +
tests/ovn.at| 230 +++-
2 files changed, 122 insertions(+), 112 deletions(-)
diff --git
Xavier Simonart (2):
tests: Make test "action parsing" support expansion.
tests: Remove table numbers from test action parsing.
tests/ovn-macros.at | 4 +
tests/ovn.at| 662 ++--
2 files changed, 338 insertions(+), 328 deletions(-)
--
On Tue, Feb 6, 2024 at 2:11 PM Aaron Conole wrote:
>
> The ovs module allows for some actions to recursively contain an action
> list for complex scenarios, such as sampling, checking lengths, etc.
> When these actions are copied into the internal flow table, they are
> evaluated to validate that
On Tue, Feb 6, 2024 at 2:31 PM 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 6 Feb 2024, at 14:46, Ilya Maximets wrote:
> On 2/6/24 14:30, Eelco Chaudron wrote:
>> Previously, the ability to override the system default for the number
>> of handler threads was broken since to the introduction of the per-CPU
>> upcall handlers.
>
> It wasn't broken, it was
On 2/6/24 14:30, Eelco Chaudron wrote:
> Previously, the ability to override the system default for the number
> of handler threads was broken since to the introduction of the per-CPU
> upcall handlers.
It wasn't broken, it was intentionally disabled. I don't think we should
introduce ability to
On 06/02/2024 12:28, Ilya Maximets wrote:
> On 2/2/24 13:46, Kevin Traynor wrote:
>> Branches 2.17/3.0/3.1/3.2 are using newer DPDK LTS releases.
>>
>> Update the faq.
>>
>> Signed-off-by: Kevin Traynor
>> ---
>> Documentation/faq/releases.rst | 8
>> 1 file changed, 4 insertions(+), 4
Previously, the ability to override the system default for the number
of handler threads was broken since to the introduction of the per-CPU
upcall handlers.
This patch rectifies this behavior by re-enabling manual configuration
of the handler thread count.
Fixes: b1e517bd2f81 ("dpif-netlink:
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.
On 2/5/24 06:10, Mike Pattrick wrote:
> Previously when an empty mutation was used to count the number of rows
> in a table, OVSDB would iterate over all rows twice. First to perform an
> RBAC check, and then to perform the no-operation.
>
> This change adds a short circuit to mutate operations
On 6 Feb 2024, at 13:58, Ilya Maximets wrote:
> On 2/6/24 13:50, Eelco Chaudron wrote:
>>
>>
>> On 6 Feb 2024, at 13:46, Ilya Maximets wrote:
>>
>>> On 2/4/24 15:40, Timothy Redaelli wrote:
A log message like this one:
Add a test case into the netlink checks that will show the number of
nested action recursions won't exceed 16. Going to 17 on a small
clone call isn't enough to exhaust the stack on (most) systems, so
it should be safe to run even on systems that don't have the fix
applied.
Signed-off-by: Aaron
The ovs module allows for some actions to recursively contain an action
list for complex scenarios, such as sampling, checking lengths, etc.
When these actions are copied into the internal flow table, they are
evaluated to validate that such actions make sense, and these calls
happen recursively.
Open vSwitch module accepts actions as a list from the netlink socket
and then creates a copy which it uses in the action set processing.
During processing of the action list on a packet, the module keeps a
count of the execution depth and exits processing if the action depth
goes too high.
On 2/6/24 13:50, Eelco Chaudron wrote:
>
>
> On 6 Feb 2024, at 13:46, Ilya Maximets wrote:
>
>> On 2/4/24 15:40, Timothy Redaelli wrote:
>>> A log message like this one:
>>>
>>> 2024-01-09T06:45:17.201Z|00071|bfd(handler2)|INFO|ovn-0af536-0: BFD state
>>> change: down->up "Neighbor Signaled
On 1/30/24 23:14, Mike Pattrick wrote:
> This patch fixes the correctness of dp_packet_inner_l4_size() when
> checking for the existence of an inner L4 header. Previously it checked
> for the outer L4 header.
>
> This function is currently only used when a packet is already flagged
> for
On 6 Feb 2024, at 13:46, Ilya Maximets wrote:
> On 2/4/24 15:40, Timothy Redaelli wrote:
>> A log message like this one:
>>
>> 2024-01-09T06:45:17.201Z|00071|bfd(handler2)|INFO|ovn-0af536-0: BFD state
>> change: down->up "Neighbor Signaled Session Down"->"Neighbor Signaled Session
>> Down".
>>
On 6 Feb 2024, at 13:45, Kevin Traynor wrote:
> On 05/02/2024 11:32, Eelco Chaudron wrote:
>>
>>
>> On 2 Feb 2024, at 13:46, Kevin Traynor wrote:
>>
>>> Update the CI and docs to use DPDK 21.11.6.
>>>
>>> Signed-off-by: Kevin Traynor
>>
>> Thanks for getting all DPDK packages upgraded. The
On 2/4/24 15:40, Timothy Redaelli wrote:
> A log message like this one:
>
> 2024-01-09T06:45:17.201Z|00071|bfd(handler2)|INFO|ovn-0af536-0: BFD state
> change: down->up "Neighbor Signaled Session Down"->"Neighbor Signaled Session
> Down".
>
> can be hard to read since '->' usually represents a
On 05/02/2024 11:32, Eelco Chaudron wrote:
>
>
> On 2 Feb 2024, at 13:46, Kevin Traynor wrote:
>
>> Update the CI and docs to use DPDK 21.11.6.
>>
>> Signed-off-by: Kevin Traynor
>
> Thanks for getting all DPDK packages upgraded. The changes look good to me.
> Did not run the GitHub actions,
On 2/2/24 13:46, Kevin Traynor wrote:
> Branches 2.17/3.0/3.1/3.2 are using newer DPDK LTS releases.
>
> Update the faq.
>
> Signed-off-by: Kevin Traynor
> ---
> Documentation/faq/releases.rst | 8
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git
Bleep bloop. Greetings Ales Musil, 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 lacks whitespace around operator
#281 FILE: utilities/ovn-debug.c:87:
Add macro for OFTABLE_CHK_IN_PORT_SEC and replace all table=73
occurrences in OF with table=OFTABLE_CHK_IN_PORT_SEC.
Signed-off-by: Ales Musil
---
tests/ovn-macros.at | 1 +
tests/ovn.at| 112 ++--
2 files changed, 58 insertions(+), 55
Add macro for OFTABLE_CT_SNAT_HAIRPIN and replace all table=70
occurrences in OF with table=OFTABLE_CT_SNAT_HAIRPIN.
Signed-off-by: Ales Musil
---
tests/ovn-macros.at | 1 +
tests/ovn.at| 206 ++--
2 files changed, 104 insertions(+), 103
On 2/5/24 23:53, Han Zhou wrote:
> For Dumitru's proposal:
>>> Maybe it would be an idea to integrate some of Han's performance testing
>>> scripts into the set of tests we already have in the upstream repo,
>>> ovn-performance.at [0], and run those in GitHub actions too.
>>> [0]
Add macro for OFTABLE_CHK_IN_PORT_SEC_ND and replace all table=74
occurrences in OF with table=OFTABLE_CHK_IN_PORT_SEC_ND.
Signed-off-by: Ales Musil
---
tests/ovn-macros.at | 1 +
tests/ovn.at| 151 ++--
2 files changed, 77 insertions(+), 75
Add macro for OFTABLE_MAC_LOOKUP and replace all table=67
occurrences in OF with table=OFTABLE_MAC_LOOKUP.
Signed-off-by: Ales Musil
---
tests/ovn-macros.at | 1 +
tests/ovn.at| 26 +-
2 files changed, 14 insertions(+), 13 deletions(-)
diff --git
Bleep bloop. Greetings Ales Musil, 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: Comment with 'xxx' marker
#241 FILE: controller/pinctrl.c:6422:
/* XXX: This handler
Add ovn-debug binary tool that can be extended with commands that
might be useful for tests/debugging of OVN environment. Currently
the tool supports only two commands:
1) "lflow-stage-to-ltable STAGE_NAME" that converts stage name into
logical flow table id.
2) "lflow-stage-to-oftable
Add macro for OFTABLE_MAC_CACHE_USE and replace all table=79
occurrences in OF with table=OFTABLE_MAC_CACHE_USE.
Signed-off-by: Ales Musil
---
tests/ovn-macros.at | 1 +
tests/ovn.at| 12 ++--
2 files changed, 7 insertions(+), 6 deletions(-)
diff --git a/tests/ovn-macros.at
Use the ovn-debug commands to determine OpenFlow table numbers
based on stage name. With this there is no need to hardcode them
and it should be future proof for stage shifts/updates.
Signed-off-by: Ales Musil
---
tests/ovn-controller.at | 342 +++---
tests/ovn.at
Add macro for OFTABLE_CHK_OUT_PORT_SEC and replace all table=75
occurrences in OF with table=OFTABLE_CHK_OUT_PORT_SEC.
Signed-off-by: Ales Musil
---
tests/ovn-macros.at | 1 +
tests/ovn.at| 121 ++--
2 files changed, 62 insertions(+), 60
Add macro for OFTABLE_ECMP_NH_MAC and replace all table=76
occurrences in OF with table=OFTABLE_ECMP_NH_MAC.
Signed-off-by: Ales Musil
---
tests/ovn-macros.at | 1 +
tests/system-ovn.at | 4 ++--
2 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/tests/ovn-macros.at
Add macro for OFTABLE_CHK_LB_AFFINITY and replace all table=78
occurrences in OF with table=OFTABLE_CHK_LB_AFFINITY.
Signed-off-by: Ales Musil
---
tests/ovn-macros.at | 1 +
tests/system-ovn-kmod.at | 16
tests/system-ovn.at | 6 +++---
3 files changed, 12
To avoid issues with hardcoded table numbers in future add rule
into check patch. The rule is only warning because there are still
legitimate use cases and not everything can be abstracted.
Signed-off-by: Ales Musil
---
utilities/checkpatch.py | 12
1 file changed, 12 insertions(+)
Add macro for OFTABLE_CHK_LB_HAIRPIN and replace all table=68
occurrences in OF with table=OFTABLE_CHK_LB_HAIRPIN.
Signed-off-by: Ales Musil
---
tests/ovn-macros.at | 1 +
tests/ovn.at| 316 ++--
2 files changed, 159 insertions(+), 158
Add macro for OFTABLE_CHK_LB_HAIRPIN_REPLY and replace all table=69
occurrences in OF with table=OFTABLE_CHK_LB_HAIRPIN_REPLY.
Signed-off-by: Ales Musil
---
tests/ovn-macros.at | 1 +
tests/ovn.at| 404 ++--
2 files changed, 203 insertions(+),
Add macro for OFTABLE_GET_FDB and replace all table=71
occurrences in OF with table=OFTABLE_GET_FDB.
Signed-off-by: Ales Musil
---
tests/ovn-macros.at | 1 +
tests/ovn.at| 16
2 files changed, 9 insertions(+), 8 deletions(-)
diff --git a/tests/ovn-macros.at
Add macro for OFTABLE_ECMP_NH and replace all table=77
occurrences in OF with table=OFTABLE_ECMP_NH.
Signed-off-by: Ales Musil
---
tests/ovn-macros.at | 1 +
tests/system-ovn.at | 4 ++--
2 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/tests/ovn-macros.at b/tests/ovn-macros.at
Add macro for OFTABLE_MAC_BINDING and replace all table=66
occurrences in OF with table=OFTABLE_MAC_BINDING.
Signed-off-by: Ales Musil
---
tests/ovn-macros.at | 1 +
tests/ovn.at| 12 ++--
2 files changed, 7 insertions(+), 6 deletions(-)
diff --git a/tests/ovn-macros.at
Add macro for OFTABLE_LOOKUP_FDB and replace all table=72
occurrences in OF with table=OFTABLE_LOOKUP_FDB.
Signed-off-by: Ales Musil
---
tests/ovn-macros.at | 1 +
tests/ovn.at| 18 +-
2 files changed, 10 insertions(+), 9 deletions(-)
diff --git a/tests/ovn-macros.at
Add macro for OFTABLE_LOG_TO_PHY and replace all table=65
occurrences in OF with table=OFTABLE_LOG_TO_PHY.
Signed-off-by: Ales Musil
---
tests/ovn-macros.at | 1 +
tests/ovn.at| 28 ++--
2 files changed, 15 insertions(+), 14 deletions(-)
diff --git
Add macro for OFTABLE_REMOTE_OUTPUT and replace all table=39
occurrences in OF with table=OFTABLE_REMOTE_OUTPUT.
Signed-off-by: Ales Musil
---
tests/ovn-macros.at | 1 +
tests/ovn.at| 40
2 files changed, 21 insertions(+), 20 deletions(-)
diff
Add macro for OFTABLE_LOCAL_OUTPUT and replace all table=40
occurrences in OF with table=OFTABLE_LOCAL_OUTPUT.
Signed-off-by: Ales Musil
---
tests/ovn-controller.at | 6 +++---
tests/ovn-macros.at | 1 +
tests/ovn.at| 24
3 files changed, 16
Add macro for OFTABLE_OUTPUT_LARGE_PKT_DETECT and replace all table=37
occurrences in OF with table=OFTABLE_OUTPUT_LARGE_PKT_DETECT.
Signed-off-by: Ales Musil
---
tests/ovn-macros.at | 1 +
tests/ovn.at| 14 +++---
2 files changed, 8 insertions(+), 7 deletions(-)
diff --git
Add macro for OFTABLE_PHY_TO_LOG and replace all table=0
occurrences in OF with table=OFTABLE_PHY_TO_LOG.
Signed-off-by: Ales Musil
---
tests/ovn-controller-vtep.at | 10 ++---
tests/ovn-controller.at | 6 +--
tests/ovn-macros.at | 2 +
tests/ovn.at | 72
Add macro for OFTABLE_OUTPUT_LARGE_PKT_PROCESS and replace all table=38
occurrences in OF with table=OFTABLE_OUTPUT_LARGE_PKT_PROCESS.
Signed-off-by: Ales Musil
---
tests/ovn-macros.at | 1 +
tests/ovn.at| 2 +-
2 files changed, 2 insertions(+), 1 deletion(-)
diff --git
Add macro for OFTABLE_LOG_INGRESS_PIPELINE and replace all table=8
occurrences in OF with table=OFTABLE_LOG_INGRESS_PIPELINE.
Signed-off-by: Ales Musil
---
tests/ovn-controller-vtep.at | 4 ++--
tests/ovn-macros.at | 1 +
tests/ovn.at | 2 +-
3 files changed, 4
Replace all usage of "sed 's/table=./table=?/'" or
"sed 's/table=../table=??/' with ovn_strip_lflows.
Adjust all table=? to table=?? and re-sort the expected flows.
Signed-off-by: Ales Musil
---
tests/ovn-northd.at | 201 ++--
tests/ovn.at| 24
Replace all usage of "sort | sed 's/table=./table=?/'" or
'sort | sed 's/table=../table=??/' with ovn_strip_lflows.
Adjust all table=? to table=?? and re-sort the expected flows.
Signed-off-by: Ales Musil
---
tests/ovn-northd.at | 260 ++--
tests/ovn.at
Replace all usage of "sed 's/table=./table=?/' | sort"
with ovn_strip_lflows and adjust all table=? to be table=??.
Signed-off-by: Ales Musil
---
tests/ovn-northd.at | 564 ++--
1 file changed, 281 insertions(+), 283 deletions(-)
diff --git
A lot of tests were relying on exact table numbers for
flow comparison either logical or OpenFlows.
To solve this problem logical flows will have stripped table
maening that the table= will become table=??. This makes it
future proof for any logical table shifts and there is no need to deal
with
On Mon, Feb 5, 2024 at 11:03 PM Dumitru Ceara wrote:
> On 12/14/23 16:29, Ales Musil wrote:
> > The fields that are not directly supported by OvS were encoded
> > via additional controller action that changed the required value.
> > This was most notably needed for ICMP need frag messages.
> >
>
The fields that are not directly supported by OvS were encoded
via additional controller action that changed the required value.
This was most notably needed for ICMP need frag messages.
Encode the field value loads as note action instead. This allows
us to find the note and act accordingly in
78 matches
Mail list logo
