Introduce check_pkt_larger action for ingress traffic
entering the cluster from a distributed gw router port
or from a gw router. This patch enables pMTU discovery
for ingress traffic.
Signed-off-by: Lorenzo Bianconi
---
northd/ovn-northd.8.xml | 60 +--
northd/ovn-northd.c |
As it is already done for distributed gw router scenario, introduce
check_pkt_larger logical flows for gw router use case.
Co-authored-by: Numan Siddique
Signed-off-by: Numan Siddique
Signed-off-by: Lorenzo Bianconi
---
northd/ovn-northd.8.xml | 24
northd/ovn-northd.c | 31
In the current codebase, check_pkt_larger is applied just for traffic
leaving the ovn cluster. This series introduces the same capability
for traffic entering the network from a gateway router or distributed
gateway router port in order to send an ICMP error packet if the frame
size is greater
Introduce build_check_pkt_len_flows_for_lrp routine to configure
check_pkt_larger logical flow for a given logical port. This is a
preliminary patch to enable check_pkt_larger support for gw router
use case.
Acked-by: Mark Michelson
Signed-off-by: Lorenzo Bianconi
---
northd/ovn-northd.c | 181
Vasu, I reviewed your v7, but added my comments in the v6 email :(
As only the documentation updated in v7, it should reflect the same code areas.
Cheers,
Eelco
On 23 Jun 2021, at 16:41, Eelco Chaudron wrote:
>> On 12 Jun 2021, at 4:09, Vasu Dasari wrote:
>
> See my inline comments below.
>
> This patch introduces the mfex function pointers which allows
> the user to switch between different miniflow extract implementations
> which are provided by the OVS based on optimized ISA CPU.
Thanks for the patch Amber/harry. Comments inline below.
>
> The user can query for the available
On 6/23/21 5:18 PM, Ferriter, Cian wrote:
>> -Original Message-
>> From: dev On Behalf Of Ferriter, Cian
>> Sent: Wednesday 23 June 2021 13:38
>> To: Ilya Maximets ; Eli Britstein ;
>> d...@openvswitch.org
>> Cc: ivan.ma...@oktetlabs.ru; Ameer Mahagneh ; Majd
>> Dibbiny
>> Subject: Re:
On 6/23/21 2:12 PM, Timothy Redaelli wrote:
> Currently, on Linux, if you try to create a system datapath called
> "bonding_masters", when you have bonding module loaded, you have a
> kernel trace
> ("sysfs: cannot create duplicate filename '/class/net/bonding_masters'").
>
> This trace appears
On Wed, Jun 23, 2021 at 1:01 AM Paolo Valerio wrote:
>
> Paolo Valerio writes:
>
> > conntrack_l4csum_err gets incremented only when corrupted icmp
> > pass through conntrack.
> > Increase it for the remaining bad checksum cases including when
> > checksum is offloaded.
> >
> > Signed-off-by:
On Wed, Jun 09, 2021 at 11:52:07AM +0200, Simon Horman wrote:
> Hi,
>
> this short test adds support for add port-based ingress policing based
> packet-per-second rate-limiting. This builds on existing support for
> byte-per-second rate limiting.
>
> Changes since v2
>
> * Remove the for loop
> On 12 Jun 2021, at 4:09, Vasu Dasari wrote:
See my inline comments below.
Cheers,
Eelco
> Currently there is an option to add/flush/show ARP/ND neighbor. This covers L3
> side. For L2 side, there is only fdb show command. This patch gives an option
> to add/del an fdb entry via ovs-appctl.
> -Original Message-
> From: dev On Behalf Of Ferriter, Cian
> Sent: Wednesday 23 June 2021 13:38
> To: Ilya Maximets ; Eli Britstein ;
> d...@openvswitch.org
> Cc: ivan.ma...@oktetlabs.ru; Ameer Mahagneh ; Majd Dibbiny
>
> Subject: Re: [ovs-dev] [PATCH V6 00/13] Netdev vxlan-decap
Acked-by: Mark Michelson
On 6/22/21 4:10 PM, num...@ovn.org wrote:
From: Numan Siddique
'pflow_output_data' should be of type 'struct ed_type_pflow_output'
and not 'struct ed_type_lflow_output'.
Fixes: e07e397b7ae("ovn-controller: Split logical flow and physical flow
processing.")
On 26 May 2021, at 14:34, Eelco Chaudron wrote:
On 18 May 2021, at 20:15, Ilya Maximets wrote:
On 5/18/21 4:44 PM, Eelco Chaudron wrote:
Add support for the dec_ttl action. Instead of programming the
datapath with
a flow that matches the packet TTL and an IP set, use a single
dec_ttl
From: Anton Ivanov
Set a soft time limit of "raft election timer"/2 on ovsdb
processing.
This improves behaviour in large heavily loaded clusters.
While it cannot fully eliminate spurious raft elections
under heavy load, it significantly decreases their number.
Processing is (to the extent
Friendly ping
> -Original Message-
> From: wangyunjian
> Sent: Tuesday, April 27, 2021 2:42 PM
> To: d...@openvswitch.org; i.maxim...@ovn.org
> Cc: dingxiaoxiong ; wangyunjian
>
> Subject: [ovs-dev] [PATCH] bridge: fix type mismatch
>
> From: Yunjian Wang
>
> Currently the function
From: Dumitru Ceara
Change the ovn-northd implementation to set the new 'controller_meter'
field for flows that need to punt packets to ovn-controller.
Protocol packets for which CoPP is enforced when sending packets to
ovn-controller (if configured):
- ARP
- ND_NS
- ND_NA
- ND_RA
- DNS
- IGMP
From: Dumitru Ceara
Add new 'Copp' (Control plane protection) table to OVN Northbound DB:
- this stores mappings between control plane protocol names and meters
that should be used to rate limit controller-destined traffic for
those protocols.
Add new 'copp' columns to the following OVN
From: Dumitru Ceara
Acked-by: Mark D. Gray
Co-authored-by: Lorenzo Bianconi
Signed-off-by: Lorenzo Bianconi
Signed-off-by: Dumitru Ceara
---
NEWS | 1 +
1 file changed, 1 insertion(+)
diff --git a/NEWS b/NEWS
index 0da7d8f97..72585e56b 100644
--- a/NEWS
+++ b/NEWS
@@ -25,6 +25,7 @@ OVN
This series respin CoPP support introduced here [0] by Dumitru rebasing on top
of ovn master branch and adding some missing meters (e.g. bfd or acl reject).
The main goal of this series is to continue the discussion about the proposed
approach and to align on CMS APIs.
For the moment DDLog is not
From: Dumitru Ceara
Add a new 'controller_meter' column to OVN Southbound Logical_Flow
table. This stores an optional string which should correspond to
the Meter that must be used for rate limiting controller actions
generated by packets hitting the flow.
Add a new 'ofctrl_add_flow_metered'
> -Original Message-
> From: Ilya Maximets [mailto:i.maxim...@ovn.org]
> Sent: Wednesday, May 12, 2021 9:03 PM
> To: wangyunjian ; Ilya Maximets
> ; d...@openvswitch.org
> Cc: Lilijun (Jerry) ; xudingke
> ; chenchanghu
> Subject: Re: [ovs-dev] [PATCH] ovsdb-server: Add limitation for
On Fri, Jun 4, 2021 at 11:19 PM Kevin Traynor wrote:
>
> This reworks the current rxq scheduling code to break it into more
> generic and reusable pieces.
>
> The behaviour does not change from a user perspective, except the logs
> are updated to be more consistent.
>
> From an implementation
On Mon, Jun 21, 2021 at 2:52 AM Han Zhou wrote:
>
> If a lflow has an lport name in the match, but when the lflow is
> processed the port-binding is not seen by ovn-controller, the
> corresponding openflow will not be created. Later if the port-binding is
> created/monitored by ovn-controller,
On 5/12/21 11:17 AM, xiangxia.m@gmail.com wrote:
> From: Tonghao Zhang
>
> For now, ovs-vswitchd use the array of the dp_meter struct
> to store meter's data, and at most, there are only 65536
> (defined by MAX_METERS) meters that can be used. But in some
> case, for example, in the edge
Hey Kevin ,
Patch looks good to me.
Builds fine , all test cases here
http://patchwork.ozlabs.org/project/openvswitch/patch/20210316154532.127858-1-ktray...@redhat.com/
pass as well.
Some minor nits inline :
> +static bool
> +pmd_rebalance_dry_run(struct dp_netdev *dp)
> +
Currently, on Linux, if you try to create a system datapath called
"bonding_masters", when you have bonding module loaded, you have a
kernel trace
("sysfs: cannot create duplicate filename '/class/net/bonding_masters'").
This trace appears since "bonding" kernel modules creates a file called
Hi Mark,
Would it be possible to install /32 IP on the router to avoid flooding all
router ports? This can be hundreds if not thousands of ports and I think I've
seen issues with flooding at that scale before. I'd have to test it in the lab
though first.
Best Regards,
Krzysztof
On Thu, Jun
I just sent an updated version:
1. Logic fixed in the outer (remotes) loop. Inner (sessions) was mostly OK.
2. Appropriate measures have been taken to ensure that the "skip to" pointer is
always valid.
3. Tested by forcing <10ms timeouts on processing and/or mandating "skip" on
every
Bleep bloop. Greetings Lorenzo Bianconi, 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
#884 FILE: utilities/ovn-nbctl.c:430:
Hi all,
As part of rebasing our AVX512 DPIF on this patchset, I tested this patchset
with partial HWOL and I'm seeing strange behaviour.
I'll report back more detailed findings soon, just wanted to mention this here
as soon as I found the issue.
Thanks,
Cian
> -Original Message-
>
> -Original Message-
> From: Ilya Maximets
> Sent: Wednesday 23 June 2021 16:25
> To: Ferriter, Cian ; Ilya Maximets
> ; Eli Britstein
> ; d...@openvswitch.org
> Cc: ivan.ma...@oktetlabs.ru; Ameer Mahagneh ; Majd Dibbiny
>
> Subject: Re: [ovs-dev] [PATCH V6 00/13] Netdev vxlan-decap
On 6/23/2021 6:18 PM, Ferriter, Cian wrote:
External email: Use caution opening links or attachments
-Original Message-
From: dev On Behalf Of Ferriter, Cian
Sent: Wednesday 23 June 2021 13:38
To: Ilya Maximets ; Eli Britstein ;
d...@openvswitch.org
Cc: ivan.ma...@oktetlabs.ru;
In order to allow showing more debug messages, increase the rate limits.
Signed-off-by: Eli Britstein
Reviewed-by: Gaetan Rivet
Acked-by: Sriharsha Basavapatna
Tested-by: Emma Finn
Tested-by: Marko Kovacevic
Signed-off-by: Ilya Maximets
---
lib/netdev-offload-dpdk.c | 2 +-
1 file changed,
Bleep bloop. Greetings Eli Britstein, 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: Unexpected sign-offs from developers who are not authors or co-authors
or
Bleep bloop. Greetings Eli Britstein, 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: Unexpected sign-offs from developers who are not authors or co-authors
or
On Mon, Jun 21, 2021 at 06:58:37PM -0700, Han Zhou wrote:
> On Thu, Jun 17, 2021 at 12:59 PM Mark Michelson wrote:
> >
> > On 6/14/21 2:44 PM, Ben Pfaff wrote:
> > > On Fri, Jun 11, 2021 at 03:48:52PM -0700, Han Zhou wrote:
> > >> The test case fails quite often for northd-ddlog because of the
[updating Jesse's email address]
On Wed, Jun 23, 2021 at 04:48:29PM +0200, Matthias May via dev wrote:
> I'm currently fighting with issues where TCP/UDP frames that are larger than
> the MTU of a GRE tunnel are dropped.
> I'm aware of the whys and how to work around the issue, but while looking
For VXLAN offload, matches should be done on outer header for tunnel
properties as well as inner packet matches. Add a function for parsing
VXLAN tunnel matches.
Signed-off-by: Eli Britstein
Reviewed-by: Gaetan Rivet
Acked-by: Sriharsha Basavapatna
Tested-by: Emma Finn
Tested-by: Marko
As a pre-step towards tunnel offloads, introduce DPDK APIs.
Signed-off-by: Eli Britstein
Reviewed-by: Gaetan Rivet
Acked-by: Sriharsha Basavapatna
Tested-by: Emma Finn
Tested-by: Marko Kovacevic
Signed-off-by: Ilya Maximets
---
lib/netdev-dpdk.c | 112
A miss in virtual port offloads means the flow with tnl_pop was
offloaded, but not the following one. Recover the state and continue
with SW processing.
Signed-off-by: Eli Britstein
Reviewed-by: Gaetan Rivet
Acked-by: Sriharsha Basavapatna
Tested-by: Emma Finn
Tested-by: Marko Kovacevic
Vports are virtual, OVS only logical devices, so rte_flows cannot be
applied as is on them. Instead, apply the rules the physical port from
which the packet has arrived, provided by orig_in_port field.
Signed-off-by: Eli Britstein
Reviewed-by: Gaetan Rivet
Acked-by: Sriharsha Basavapatna
VXLAN decap in OVS-DPDK configuration consists of two flows:
F1: in_port(ens1f0),eth(),ipv4(),udp(), actions:tnl_pop(vxlan_sys_4789)
F2: tunnel(),in_port(vxlan_sys_4789),eth(),ipv4(), actions:ens1f0_0
F1 is a classification flow. It has outer headers matches and it
classifies the packet as a
From: Ilya Maximets
'linux_tc' flow API suitable only for tunneling vports with backing
linux interfaces. DPDK flow API is not suitable for such ports.
With this change we could drop vport restriction from dpif-netdev.
This is a prerequisite for enabling vport offloading in DPDK.
From: Ilya Maximets
Virtual interfaces like vports or dpdk vhost-user ports have no
proper ifindex, while still supporting some offloads.
This is a prerequisite for tunneling vport offloading with DPDK
flow API.
Signed-off-by: Ilya Maximets
Signed-off-by: Eli Britstein
Reviewed-by: Gaetan
Recover the packet if it was partially processed by the HW. Fallback to
lookup flow by mark association.
Signed-off-by: Eli Britstein
Reviewed-by: Gaetan Rivet
Acked-by: Sriharsha Basavapatna
Tested-by: Emma Finn
Tested-by: Marko Kovacevic
Signed-off-by: Ilya Maximets
---
lib/dpif-netdev.c
Add the acceptance of vxlan devices to netdev_dpdk_flow_api_supported()
API, to allow offloading of DPDK vxlan devices.
Signed-off-by: Eli Britstein
Reviewed-by: Gaetan Rivet
Acked-by: Sriharsha Basavapatna
Tested-by: Emma Finn
Tested-by: Marko Kovacevic
Signed-off-by: Ilya Maximets
---
Support tunnel pop action.
Signed-off-by: Eli Britstein
Reviewed-by: Gaetan Rivet
Acked-by: Sriharsha Basavapatna
Tested-by: Emma Finn
Tested-by: Marko Kovacevic
Signed-off-by: Ilya Maximets
---
Documentation/howto/dpdk.rst | 1 +
NEWS | 2 +
When the HW offload involves multiple flows, like in tunnel decap path,
it is possible that not all flows in the path are offloaded, resulting
in partial processing in HW. In order to proceed with rest of the
processing in SW, the packet state has to be recovered as if it was
processed in SW from
From: Sriharsha Basavapatna
When an encapsulated packet is recirculated through a TUNNEL_POP
action, the metadata gets reinitialized and the originating physical
port information is lost. When this flow gets processed by the vport
and it needs to be offloaded, we can't figure out the physical
Introduce an API to traverse the ports added to the offload ports map,
with a generic callback for each one.
Signed-off-by: Eli Britstein
Reviewed-by: Gaetan Rivet
Acked-by: Sriharsha Basavapatna
Tested-by: Emma Finn
Tested-by: Marko Kovacevic
Signed-off-by: Ilya Maximets
---
Bleep bloop. Greetings Eli Britstein, 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: Unexpected sign-offs from developers who are not authors or co-authors
or
Bleep bloop. Greetings Eli Britstein, 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: Unexpected sign-offs from developers who are not authors or co-authors
or
On Wed, Jun 23, 2021 at 11:26 AM Mark Michelson wrote:
>
> Acked-by: Mark Michelson
Thanks. Applied to the main branch.
Numan
>
> On 6/22/21 4:10 PM, num...@ovn.org wrote:
> > From: Numan Siddique
> >
> > 'pflow_output_data' should be of type 'struct ed_type_pflow_output'
> > and not
Hi Jesse, Hi List
I'm currently fighting with issues where TCP/UDP frames that are larger than
the MTU of a GRE tunnel are dropped.
I'm aware of the whys and how to work around the issue, but while looking for
solutions i stumbled over the fact that:
* [1] added PMTUD support to OVS
* [2]
I don't see the charter, etc. in the OVN tree. I know we started the
process but I don't know whether we finished it.
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
Bleep bloop. Greetings Eli Britstein, 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: Unexpected sign-offs from developers who are not authors or co-authors
or
Bleep bloop. Greetings Eli Britstein, 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: Unexpected sign-offs from developers who are not authors or co-authors
or
Bleep bloop. Greetings Eli Britstein, 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: Unexpected sign-offs from developers who are not authors or co-authors
or
Bleep bloop. Greetings Eli Britstein, 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: Unexpected sign-offs from developers who are not authors or co-authors
or
Bleep bloop. Greetings Eli Britstein, 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: Unexpected sign-offs from developers who are not authors or co-authors
or
On Wed, 2021-06-23 at 10:06 -0700, Ben Pfaff wrote:
> [updating Jesse's email address]
>
> On Wed, Jun 23, 2021 at 04:48:29PM +0200, Matthias May via dev wrote:
> > I'm currently fighting with issues where TCP/UDP frames that are
> > larger than the MTU of a GRE tunnel are dropped.
> > I'm aware
Hey Kevin ,
Patch looks good to me.
Builds fine , all test cases here
http://patchwork.ozlabs.org/project/openvswitch/patch/20210316154532.127858-1-ktray...@redhat.com/
pass as well.
The groups assignment works fine too.
>From vswitchd logs:
dpif_netdev|INFO|PMD auto load balance load
Bleep bloop. Greetings Eli Britstein, 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: Unexpected sign-offs from developers who are not authors or co-authors
or
Bleep bloop. Greetings Eli Britstein, 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: Unexpected sign-offs from developers who are not authors or co-authors
or
Bleep bloop. Greetings Eli Britstein, 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: Unexpected sign-offs from developers who are not authors or co-authors
or
Bleep bloop. Greetings Eli Britstein, 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: Unexpected sign-offs from developers who are not authors or co-authors
or
On Thu, Jun 17, 2021 at 05:18:18PM +0100, Cian Ferriter wrote:
> From: Harry van Haaren
>
> This commit adds a new command to allow the user to switch
> the active DPIF implementation at runtime. A probe function
> is executed before switching the DPIF implementation, to ensure
> the CPU is
On 6/23/21 5:52 PM, Eli Britstein wrote:
> VXLAN decap in OVS-DPDK configuration consists of two flows:
> F1: in_port(ens1f0),eth(),ipv4(),udp(), actions:tnl_pop(vxlan_sys_4789)
> F2: tunnel(),in_port(vxlan_sys_4789),eth(),ipv4(), actions:ens1f0_0
>
> F1 is a classification flow. It has outer
On 3/16/21 4:45 PM, Kevin Traynor wrote:
> These tests focus on enabling/disabling and user parameters.
>
> Co-Authored-by: David Marchand
> Signed-off-by: David Marchand
> Signed-off-by: Kevin Traynor
>
> ---
> v2:
> - Remove above max documented interval test
> - Add David's code to combine
Currently, if you try to set subtable-lookup-prio-set when you don't have
any datapath (for example if an user wants to set AVX512 before creating
any bridge) it sets it globally (dpcls_subtable_set_prio),
but it returns an error:
please specify an existing datapath
ovs-appctl: ovs-vswitchd:
On 6/23/21 8:54 PM, Timothy Redaelli wrote:
> Currently, if you try to set subtable-lookup-prio-set when you don't have
> any datapath (for example if an user wants to set AVX512 before creating
> any bridge) it sets it globally (dpcls_subtable_set_prio),
> but it returns an error:
>
> please
On Thu, Jun 17, 2021 at 05:18:19PM +0100, Cian Ferriter wrote:
> From: Harry van Haaren
>
> This commit adds a new command to retrieve the list of available
> DPIF implementations. This can be used by to check what implementations
> of the DPIF are available in any given OVS binary.
>
> Usage:
On Wed, Jun 23, 2021 at 03:21:37PM -0400, Numan Siddique wrote:
> On Wed, Jun 23, 2021 at 1:58 PM Ben Pfaff wrote:
> >
> > I don't see the charter, etc. in the OVN tree. I know we started the
> > process but I don't know whether we finished it.
>
> I'm not sure about the official status but I
Bleep bloop. Greetings Timothy Redaelli, 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: Improper whitespace around control block
#79 FILE: lib/dpif-netdev.c:1374:
On 6/23/21 5:48 PM, Eli Britstein wrote:
@@ -1009,7 +1009,7 @@ parse_vxlan_match(struct flow_patterns *patterns,
return 0;
}
-static int
+static int OVS_UNUSED
>
> Note that if experimental is allowed, the OVS_UNUSED attribute is misleading.
Yeah, I know.
On Wed, Jun 23, 2021 at 1:58 PM Ben Pfaff wrote:
>
> I don't see the charter, etc. in the OVN tree. I know we started the
> process but I don't know whether we finished it.
I'm not sure about the official status but I presumed that OVN is part of LF.
Trishan de Lanerolle (CC'ed) has reached
Currently, if you try to set subtable-lookup-prio-set when you don't have
any datapath (for example if an user wants to set AVX512 before creating
any bridge) it sets it globally (dpcls_subtable_set_prio),
but it returns an error:
please specify an existing datapath
ovs-appctl: ovs-vswitchd:
On Wed, Jun 23, 2021 at 7:48 AM Numan Siddique wrote:
>
> On Mon, Jun 21, 2021 at 2:52 AM Han Zhou wrote:
> >
> > If a lflow has an lport name in the match, but when the lflow is
> > processed the port-binding is not seen by ovn-controller, the
> > corresponding openflow will not be created.
This allows L3+ ACLs to match against double tagged vlan traffic on
vlan-passthru switches.
The default in OVS is vlan-limit=1 for backwards compatibility. This
means packets are not "parsed" deeper than one tag level.
This patch sets it to 0, which means "parse as deep as OVS supports".
Right
On Wed, 2021-06-09 at 14:11 -0500, Dan Williams wrote:
> Signed-off-by: Dan Williams
> ---
> v3: fix line wrapping
> v2: put --election-timer arg before create-cluster per Ilya
Ping on this patch?
Thanks!
Dan
>
> utilities/ovs-lib.in | 10 --
> 1 file changed, 8 insertions(+), 2
On Thu, Jun 17, 2021 at 05:18:22PM +0100, Cian Ferriter wrote:
> From: Harry van Haaren
>
> This commit implements larger subtable searches in avx512. A limitation
> of the previous implementation was that up to 8 blocks of miniflow
> data could be matched on (so a subtable with 8 blocks was
On Thu, Jun 17, 2021 at 05:18:17PM +0100, Cian Ferriter wrote:
> From: Harry van Haaren
>
> This commit adds the AVX512 implementation of DPIF functionality,
> specifically the dp_netdev_input_outer_avx512 function. This function
> only handles outer (no re-circulations), and is optimized to use
On Thu, Jun 17, 2021 at 05:18:24PM +0100, Cian Ferriter wrote:
> From: Harry van Haaren
>
> As a small optimization, this patch caches the result of a CPU ISA
> check from DPDK. Particularly in the case of running the DPCLS
> autovalidator (which repeatedly probes subtables) this reduces
> the
On Thu, Jun 17, 2021 at 05:18:21PM +0100, Cian Ferriter wrote:
> Signed-off-by: Cian Ferriter
>
> ---
Acked-by: Flavio Leitner
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
On Thu, Jun 17, 2021 at 05:18:23PM +0100, Cian Ferriter wrote:
> From: Harry van Haaren
>
> This commit adds more subtables to be specialized. The traffic
> pattern here being matched is VXLAN traffic subtables, which commonly
> have (5,3), (9,1) and (9,4) subtable fingerprints.
>
>
On Thu, Jun 17, 2021 at 05:18:20PM +0100, Cian Ferriter wrote:
> It is possible for packets traversing the userspace datapath to match a
> flow before hitting on EMC by using a mark ID provided by a NIC. Add a
> PMD statistic for this hit.
>
> Signed-off-by: Cian Ferriter
>
> ---
Acked-by:
On Thu, Jun 17, 2021 at 05:18:25PM +0100, Cian Ferriter wrote:
> From: Harry van Haaren
>
> This commit enables the AVX512-VPOPCNTDQ Vector Popcount
> instruction. This instruction is not available on every CPU
> that supports the AVX512-F Foundation ISA, hence it is enabled
> only when the
88 matches
Mail list logo
