Re: [ovs-dev] [PATCH v10 1/4] mfex_avx512: Calculate pkt offsets at compile time.

> -Original Message- > From: Amber, Kumar > Sent: Tuesday, May 31, 2022 3:01 PM > To: ovs-dev@openvswitch.org > Cc: echau...@redhat.com; i.maxim...@ovn.org; Ferriter, Cian > ; Stokes, Ian ; > david.march...@redhat.com; f...@sysclose.org; Van Haaren, Harry > ; Amber, Kumar > Subject:

Re: [ovs-dev] [PATCH v10 3/4] dpif-netdev/mfex: Add AVX512 ipv6 traffic profiles

> -Original Message- > From: Amber, Kumar > Sent: Tuesday 31 May 2022 15:01 > To: ovs-dev@openvswitch.org > Cc: echau...@redhat.com; i.maxim...@ovn.org; Ferriter, Cian > ; Stokes, Ian > ; david.march...@redhat.com; f...@sysclose.org; Van > Haaren, Harry > ; Amber, Kumar > Subject:

Re: [ovs-dev] [PATCH] netdev-linux: skip some internal kernel stats gathering

> On May 26, 2022, at 9:11 PM, Jon Kohler wrote: > > For netdev_linux_update_via_netlink(), hint to the kernel that > we do not need it to gather netlink internal stats when we want > to update the netlink flags, as those stats are not rendered > within OVS. > > Background: > ovs-vswitchd

Re: [ovs-dev] [RFC PATCH 0/6] Remove OVS kernel driver

On 5/25/2022 6:47 AM, Flavio Leitner wrote: Hi Greg, On Mon, May 23, 2022 at 09:10:36PM +0200, Ilya Maximets wrote: On 5/19/22 20:04, Gregory Rose wrote: On 4/15/2022 2:42 PM, Greg Rose wrote: It is time to remove support for the OVS kernel driver and push towards use of the upstream

[ovs-dev] [PATCH v10 3/4] dpif-netdev/mfex: Add AVX512 ipv6 traffic profiles

Add AVX512 Ipv6 optimized profile for vlan/IPv6/UDP and vlan/IPv6/TCP, IPv6/UDP and IPv6/TCP. MFEX autovalidaton test-case already has the IPv6 support for validating against the scalar mfex. Signed-off-by: Kumar Amber Signed-off-by: Harry van Haaren Co-authored-by: Harry van Haaren --- v10:

[ovs-dev] [PATCH v10 0/4] MFEX Optimizations IPv6 + Hashing Optimizations

The patchset introuduces IPv6 optimized MFEX profiles with AVX512 which can deliver upto 20% to 30% gain in performance over the existing scalar data-path. Hashing Optimization are also included which can further improve performance by approximately 10%. The patch also removes the magic numbers

[ovs-dev] [PATCH v10 1/4] mfex_avx512: Calculate pkt offsets at compile time.

The patch removes magic numbers pkt offsets and minimum packet lenght and instead calculate it at compile time. Signed-off-by: Kumar Amber --- v8: - Rename offset defines. --- --- lib/dpif-netdev-extract-avx512.c | 29 + 1 file changed, 21 insertions(+), 8

[ovs-dev] [PATCH v10 2/4] mfex_avx512: Calculate miniflow_bits at compile time.

The patch removes magic numbers from miniflow_bits and calculates the bits at compile time. This also makes it easier to handle any ABI changes. Signed-off-by: Kumar Amber --- lib/dpif-netdev-extract-avx512.c | 21 + 1 file changed, 17 insertions(+), 4 deletions(-) diff

[ovs-dev] [PATCH v10 4/4] dpif-netdev/mfex: Add ipv6 profile based hashing.

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 IPv6 profile specific hashing which uses fixed offsets into the packet to improve hashing performance. Signed-off-by: Kumar Amber

[ovs-dev] [PATCH] odp-util: Ignore unknown attributes in parse_key_and_mask_to_match()

When processing netlink messages, we should ignore unknown OVS_KEY_ATTR as we can assume if newer attributes are present, they are backward compatible. This patch also updates the existing comments to better explain what happens in the error cases. At this place in the code, we can not ignore the

Re: [ovs-dev] [PATCH net-next v2] net: openvswitch: fix uAPI incompatibility with existing user space

On 23 May 2022, at 14:54, Eelco Chaudron wrote: > On 17 May 2022, at 13:10, Eelco Chaudron wrote: > >> On 12 May 2022, at 12:08, Vlad Buslov wrote: >> >>> On Thu 12 May 2022 at 12:19, Eelco Chaudron wrote: On 7 Apr 2022, at 12:22, Ilya Maximets wrote: > On 4/7/22 10:02, Vlad

Re: [ovs-dev] [PATCH v10 2/4] mfex_avx512: Calculate miniflow_bits at compile time.

> -Original Message- > From: Amber, Kumar > Sent: Tuesday, May 31, 2022 3:01 PM > To: ovs-dev@openvswitch.org > Cc: echau...@redhat.com; i.maxim...@ovn.org; Ferriter, Cian > ; Stokes, Ian ; > david.march...@redhat.com; f...@sysclose.org; Van Haaren, Harry > ; Amber, Kumar > Subject:

[ovs-dev] [PATCH ovn v2] OVN-CI: Add test cases with monitor-all enabled.

currently ovn ci only has one test case with the option ovn-monitor-all enabled, this patch will add one more execution with option ovn-monitor-all=true for each test case that are wrapped by OVN_FOR_EACH_NORTHD macro. This will more or less double the number of test cases. It is possible to

Re: [ovs-dev] [ovs-dev v7 1/3] ovs-rcu: Add ovsrcu_barrier

On 5/27/22 17:03, Aaron Conole wrote: > Peng He writes: > >> ovsrcu_barrier will block the current thread until all the postponed >> rcu job has been finished. it's like a OVS version of the Linux >> kernel rcu_barrier(). >> >> Signed-off-by: Peng He >> Co-authored-by: Eelco Chaudron >>

[ovs-dev] [PATCH v9 2/4] mfex_avx512: Calculate miniflow_bits at compile time.

The patch removes magic numbers from miniflow_bits and calculates the bits at compile time. This also makes it easier to handle any ABI changes. Signed-off-by: Kumar Amber --- lib/dpif-netdev-extract-avx512.c | 21 + 1 file changed, 17 insertions(+), 4 deletions(-) diff

[ovs-dev] [PATCH v9 0/4] MFEX Optimizations IPv6 + Hashing Optimizations

The patchset introuduces IPv6 optimized MFEX profiles with AVX512 which can deliver upto 20% to 30% gain in performance over the existing scalar data-path. Hashing Optimization are also included which can further improve performance by approximately 10%. The patch also removes the magic numbers

[ovs-dev] [PATCH v9 1/4] mfex_avx512: Calculate pkt offsets at compile time.

The patch removes magic numbers pkt offsets and minimum packet lenght and instead calculate it at compile time. Signed-off-by: Kumar Amber --- v8: - Rename offset defines. --- --- lib/dpif-netdev-extract-avx512.c | 29 + 1 file changed, 21 insertions(+), 8

Re: [ovs-dev] [PATCH branch-2.17] dpdk: Use DPDK 21.11.1 release

On 5/26/22 18:14, Kevin Traynor wrote: > On 19/05/2022 14:50, Michael Phelan wrote: >>    Modify ci linux build script to use the latest DPDK stable release >> 21.11.1. >>    Modify Documentation to use the latest DPDK stable release 21.11.1. >>    Update NEWS file to reflect the latest DPDK

Re: [ovs-dev] [PATCH] dpdk: Use DPDK 21.11.1 release

On 5/26/22 18:14, Kevin Traynor wrote: > Hi Michael, > > On 19/05/2022 14:49, Michael Phelan wrote: >>    Modify ci linux build script to use the latest DPDK stable release >> 21.11.1. >>    Modify Documentation to use the latest DPDK stable release 21.11.1. >>    Update NEWS file to reflect the

Re: [ovs-dev] [PATCH v2 0/2] Rxq scheduling code rework.

On 5/19/22 14:19, Kevin Traynor wrote: > Any issues with merging these? They are a dependency for cross-numa patches > Anurag/Jan are submitting, so it will be easier for reviewers and avoid CI > failure if they are in beforehand. > > On 05/04/2022 15:13, Kevin Traynor wrote: >> These are

Re: [ovs-dev] [PATCH v4 1/2] handlers: Create additional handler threads when using CPU isolation

Michael Santana writes: > Additional threads are required to service upcalls when we have CPU > isolation (in per-cpu dispatch mode). The formula used to calculate > the number of handler threads to create is as follows > > handlers_n = min(next_prime(active_cores+1), total_cores) > > Where

[ovs-dev] [PATCH ovn] OVN-CI: Add test cases with monitor-all enabled.

currently ovn ci only has one test case with the option ovn-monitor-all enabled, this patch will add one more execution with option ovn-monitor-all=true for each test case that are wrapped by OVN_FOR_EACH_NORTHD macro. This will more or less double the number of test cases. It is possible to

Re: [ovs-dev] [PATCH v5 5/5] acinclude: Add seperate checks for AVX512 ISA.

On 5/18/22 10:51, Ferriter, Cian wrote: > > >> -Original Message- >> From: Pai G, Sunil >> Sent: Tuesday 17 May 2022 15:52 >> To: Ferriter, Cian ; ovs-dev@openvswitch.org >> Cc: Van Haaren, Harry >> Subject: RE: [PATCH v5 5/5] acinclude: Add seperate checks for AVX512 ISA. >> >>>

Re: [ovs-dev] [PATCH ovn] test: Make vtep test pass more consistently.

Hi Dumitru, thanks for the review. On 5/24/22 07:17, Dumitru Ceara wrote: On 5/23/22 20:43, Mark Michelson wrote: The "vtep: 3 HVs, 1 VIFs/HV, 1 GW, 1 LS" test fails consistently on PPC. The test expects a flow with the following to be present: is_chassis_resident("cr-lrp1") ||

Re: [ovs-dev] [PATCH] raft: Don't use HMAP_FOR_EACH_SAFE when logging commands.

On 5/16/22 13:15, Dumitru Ceara wrote: > There's no need to do that because we're not changing the hmap. > Also, if DEBUG logging is disabled there's no need to iterate at > all. > > Fixes: 5a9b53a51ec9 ("ovsdb raft: Fix duplicated transaction execution when > leader failover.") > Signed-off-by:

Re: [ovs-dev] [PATCH v5 0/2] Fix ALB parameters type and value mismatch.

On 5/24/22 15:48, mit...@outlook.com wrote: > From: Lin Huang > > This series fix ALB parameters type mismatch and set a hard limit > to rebalance_intvl. > > It also includes some self-tests. > > Changes since v3: > 1. Fix CI warnings > > Changes since v2: >1. Fix type mismatch int

[ovs-dev] [PATCH v9 3/4] dpif-netdev/mfex: Add AVX512 ipv6 traffic profiles

Add AVX512 Ipv6 optimized profile for vlan/IPv6/UDP and vlan/IPv6/TCP, IPv6/UDP and IPv6/TCP. MFEX autovalidaton test-case already has the IPv6 support for validating against the scalar mfex. Signed-off-by: Kumar Amber Signed-off-by: Harry van Haaren Co-authored-by: Harry van Haaren --- v9:

[ovs-dev] [PATCH v9 4/4] dpif-netdev/mfex: Add ipv6 profile based hashing.

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 IPv6 profile specific hashing which uses fixed offsets into the packet to improve hashing performance. Signed-off-by: Kumar Amber

Re: [ovs-dev] [PATCH v9 3/4] dpif-netdev/mfex: Add AVX512 ipv6 traffic profiles

Bleep bloop. Greetings Kumar Amber, 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: Failed to merge in the changes. hint: Use 'git am --show-current-patch' to see the failed

Re: [ovs-dev] [PATCH v9 4/4] dpif-netdev/mfex: Add ipv6 profile based hashing.

Bleep bloop. Greetings Kumar Amber, 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, please

Re: [ovs-dev] [PATCH v4 2/2] handlers: Fix handlers mapping

On 5/27/22 13:18, Michael Santana wrote: The handler and CPU mapping in upcalls are incorrect, and this is specially noticeable systems with cpu isolation enabled. Say we have a 12 core system where only every even number CPU is enabled C0, C2, C4, C6, C8, C10 This means we will create an

Re: [ovs-dev] [RFC PATCH 0/6] Remove OVS kernel driver

On 5/31/22 17:36, Gregory Rose wrote: > > > On 5/25/2022 6:47 AM, Flavio Leitner wrote: >> >> Hi Greg, >> >> >> On Mon, May 23, 2022 at 09:10:36PM +0200, Ilya Maximets wrote: >>> On 5/19/22 20:04, Gregory Rose wrote: On 4/15/2022 2:42 PM, Greg Rose wrote: > It is time to

[ovs-dev] [PATCH] datapath-windows: Detect truncate action.

Truncate action is not supported in Windows datapath, but currently it incorrectly reports to ovs-vswitchd as supported blow "system@ovs-system: Datapath supports truncate action". The patches detects it and returns not support. Cc: Alin-Gabriel Serdean Cc: Wilson Peng Signed-off-by: William Tu

Re: [ovs-dev] [PATCH v4 1/2] handlers: Create additional handler threads when using CPU isolation

On 5/31/22 09:01, Aaron Conole wrote: Michael Santana writes: Additional threads are required to service upcalls when we have CPU isolation (in per-cpu dispatch mode). The formula used to calculate the number of handler threads to create is as follows handlers_n =

[ovs-dev] [PATCH] datapath-windows: Detect CT timeout.

CT timeout is not supported in Windows datapath, but currently it incorrectly reports to ovs-vswitchd as supported blow "system@ovs-system: Datapath supports timeout policy in conntrack action". The patches detects it and returns not support. Cc: Alin-Gabriel Serdean Cc: Wilson Peng

Re: [ovs-dev] [PATCH] datapath-windows: Detect truncate action.

On 5/31/22 20:27, William Tu wrote: > Truncate action is not supported in Windows datapath, but > currently it incorrectly reports to ovs-vswitchd as supported blow > "system@ovs-system: Datapath supports truncate action". > The patches detects it and returns not support. > > Cc: Alin-Gabriel

[ovs-dev] [PATCH ovn v11 3/4] Clone packets to all port chassis

When multiple chassis are set in requested-chassis, port binding is configured in multiple cluster locations. In case of live migration scenario, only one of the locations run a workload at a particular point in time. Yet, it's expected that the workload may switch to running at an additional

[ovs-dev] [PATCH ovn v11 4/4] Implement RARP activation strategy for ports

When options:activation-strategy is set to "rarp" for LSP, when used in combination with multiple chassis names listed in options:requested-chassis, additional chassis will install special flows that would block all ingress and egress traffic for the port until a special activation event happens.

[ovs-dev] [PATCH ovn v11 2/4] Support LSP:options:requested-chassis as a list

When the option is set to a comma separated list of chassis names, OVN will attempt to bind the port at any number of other locations in addition to the main chassis. This is useful in live migration scenarios where it's important to prepare the environment for workloads to move to, avoiding

[ovs-dev] [PATCH ovn v11 1/4] Update port-up on main chassis only

In a future patch, there will be a scenario where the same port has attachments at multiple (specifically, 2) chassis, so make sure that 'up' property is updated by the main chassis only. Acked-by: Numan Siddique Signed-off-by: Ihar Hrachyshka --- controller/binding.c| 9 ++---

[ovs-dev] [PATCH ovn v11 0/4] Support multiple requested-chassis

Another respin. Still the RARP activation strategy patch has pending questions to resolve. The rest are ready to merge. RARP activation strategy patch limitations, reiterated: - flows are deleted from pinctrl thread; the idea would be to make the thread inform main thread about the necessary

Re: [ovs-dev] [PATCH] datapath-windows: Detect CT timeout.

On 5/31/22 20:29, William Tu wrote: > CT timeout is not supported in Windows datapath, but > currently it incorrectly reports to ovs-vswitchd as supported blow > "system@ovs-system: Datapath supports timeout policy in conntrack > action". The patches detects it and returns not support. > > Cc:

Re: [ovs-dev] [ovs-discuss] Commit 355fef6f2 seems to break connectivity in my setup

On Mon, May 30, 2022 at 5:25 PM Frode Nordahl wrote: > > On Fri, May 27, 2022 at 10:04 PM Ilya Maximets wrote: > > > > On 5/26/22 14:53, Frode Nordahl wrote: > > > > > > > > > tor. 26. mai 2022, 14:45 skrev Ilya Maximets > > >: > > > > > > On 5/26/22 13:00, Frode

Re: [ovs-dev] [RFC PATCH 0/6] Remove OVS kernel driver

On Tue, May 31, 2022 at 7:05 PM Ilya Maximets wrote: > > On 5/31/22 17:36, Gregory Rose wrote: > > > > > > On 5/25/2022 6:47 AM, Flavio Leitner wrote: > >> > >> Hi Greg, > >> > >> > >> On Mon, May 23, 2022 at 09:10:36PM +0200, Ilya Maximets wrote: > >>> On 5/19/22 20:04, Gregory Rose wrote: >

Re: [ovs-dev] [ovs-discuss] Commit 355fef6f2 seems to break connectivity in my setup

On 5/31/22 21:15, Frode Nordahl wrote: > On Mon, May 30, 2022 at 5:25 PM Frode Nordahl > wrote: >> >> On Fri, May 27, 2022 at 10:04 PM Ilya Maximets wrote: >>> >>> On 5/26/22 14:53, Frode Nordahl wrote: tor. 26. mai 2022, 14:45 skrev Ilya Maximets >>> >:

Re: [ovs-dev] [PATCH] datapath-windows: Detect CT timeout.

On Tue, May 31, 2022 at 12:09 PM Ilya Maximets wrote: > > On 5/31/22 20:29, William Tu wrote: > > CT timeout is not supported in Windows datapath, but > > currently it incorrectly reports to ovs-vswitchd as supported blow > > "system@ovs-system: Datapath supports timeout policy in conntrack > >

Re: [ovs-dev] [PATCH] datapath-windows: Detect truncate action.

On Tue, May 31, 2022 at 12:06 PM Ilya Maximets wrote: > > On 5/31/22 20:27, William Tu wrote: > > Truncate action is not supported in Windows datapath, but > > currently it incorrectly reports to ovs-vswitchd as supported blow > > "system@ovs-system: Datapath supports truncate action". > > The