Re: [ovs-dev] [RFC PATCH 4/7] net: openvswitch: ovs_vport_receive reduce stack usage

On Fri Sep 29, 2023 at 6:38 PM AEST, Eelco Chaudron wrote: > > > On 29 Sep 2023, at 9:00, Nicholas Piggin wrote: > > > On Fri Sep 29, 2023 at 1:26 AM AEST, Aaron Conole wrote: > >> Nicholas Piggin writes: > >> > >>> Dynamically allocating the sw_flow_key reduces stack usage of > >>> ovs_vport_rece

[ovs-dev] [PATCH ovn] Don't mention packet cloning when failing to find tunnel

Packet cloning is relevant to multichassis ports only, so the message made no sense for single chassis ports with no additional_chassis. Even for multichassis ports, this part of the message is redundant and confusing. It's better to remove it. Signed-off-by: Ihar Hrachyshka --- controller/phys

[ovs-dev] [PATCH v5 2/2] ofproto-dpif-mirror: Add support for pre-selection filter.

Currently a bridge mirror will collect all packets and tools like ovs-tcpdump can apply additional filters after they have already been duplicated by vswitchd. This can result in inefficient collection. This patch adds support to apply pre-selection to bridge mirrors, which can limit which packets

[ovs-dev] [PATCH v5 1/2] ofproto-dpif-mirror: Reduce number of function parameters.

Previously the mirror_set() and mirror_get() functions took a large number of parameters, which was inefficient and difficult to read and extend. This patch moves most of the parameters into a struct. Signed-off-by: Mike Pattrick --- ofproto/ofproto-dpif-mirror.c | 61 ++-

Re: [ovs-dev] [PATCH v4 2/2] ofproto-dpif-mirror: Add support for pre-selection filter

Bleep bloop. Greetings Mike Pattrick, 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: ofproto-dpif-mirror: Add support

Re: [ovs-dev] [PATCH v4 1/2] ofproto-dpif-mirror: Reduce number of function parameters

Bleep bloop. Greetings Mike Pattrick, 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: ofproto-dpif-mirror: Reduce numbe

[ovs-dev] [PATCH v4 2/2] ofproto-dpif-mirror: Add support for pre-selection filter

Currently a bridge mirror will collect all packets and tools like ovs-tcpdump can apply additional filters after they have already been duplicated by vswitchd. This can result in inefficient collection. This patch adds support to apply pre-selection to bridge mirrors, which can limit which packets

[ovs-dev] [PATCH v4 1/2] ofproto-dpif-mirror: Reduce number of function parameters

Previously the mirror_set() and mirror_get() functions took a large number of parameters, which was inefficient and difficult to read and extend. This patch moves most of the parameters into a struct. Signed-off-by: Mike Pattrick --- ofproto/ofproto-dpif-mirror.c | 61 ++-

[ovs-dev] [PATCH ovn v2] controller: Don't artificially limit group and meter IDs to 16bit.

OVS actually supports way more. Use the real numbers instead. To avoid preallocating huge bitmaps for the group/meter IDs, switch to id-pool instead (as suggested by Ilya). Reported-at: https://issues.redhat.com/browse/FDP-70 Suggested-by: Ilya Maximets Signed-off-by: Dumitru Ceara --- V2: - Ad

Re: [ovs-dev] [PATCH ovn] controller: Don't artificially limit group and meter IDs to 16bit.

On 10/4/23 18:18, Ilya Maximets wrote: > On 10/4/23 16:40, Dumitru Ceara wrote: >> OVS actually supports way more. Use the real numbers instead. >> To avoid preallocating huge bitmaps for the group/meter IDs, switch to >> id-pool instead (as suggested by Ilya). >> >> Reported-at: https://issues.re

Re: [ovs-dev] [PATCH ovn] controller: Don't artificially limit group and meter IDs to 16bit.

On 10/4/23 16:40, Dumitru Ceara wrote: > OVS actually supports way more. Use the real numbers instead. > To avoid preallocating huge bitmaps for the group/meter IDs, switch to > id-pool instead (as suggested by Ilya). > > Reported-at: https://issues.redhat.com/browse/FDP-70 > Suggested-by: Ilya M

Re: [ovs-dev] [PATCH ovn] northd: Incrementally process SB.Load_balancer updates.

On Wed, Oct 4, 2023 at 4:41 AM Ales Musil wrote: > > On Wed, Oct 4, 2023 at 12:23 AM Dumitru Ceara wrote: > > > > Commit 9deb000536e0 ("northd: Remove potential duplicates in SB > > Load_Balancer table.") added code to northd to ensure that SB duplicates > > get removed. Quoting from that commit

Re: [ovs-dev] [PATCH] Documentation: Add CVE-2022-40982, aka Downfall reference.

Eelco Chaudron writes: > Added a reference to the DPDK documentation as a result of > analyzing the OVS code for potential performance impacts due > to the Downfall mitigation. > > Signed-off-by: Eelco Chaudron > --- Acked-by: Aaron Conole ___ dev m

Re: [ovs-dev] [RFC PATCH 4/7] net: openvswitch: ovs_vport_receive reduce stack usage

"Nicholas Piggin" writes: > On Fri Sep 29, 2023 at 1:26 AM AEST, Aaron Conole wrote: >> Nicholas Piggin writes: >> >> > Dynamically allocating the sw_flow_key reduces stack usage of >> > ovs_vport_receive from 544 bytes to 64 bytes at the cost of >> > another GFP_ATOMIC allocation in the receive

Re: [ovs-dev] [RFC PATCH 4/7] net: openvswitch: ovs_vport_receive reduce stack usage

"Nicholas Piggin" writes: > On Fri Sep 29, 2023 at 6:38 PM AEST, Eelco Chaudron wrote: >> >> >> On 29 Sep 2023, at 9:00, Nicholas Piggin wrote: >> >> > On Fri Sep 29, 2023 at 1:26 AM AEST, Aaron Conole wrote: >> >> Nicholas Piggin writes: >> >> >> >>> Dynamically allocating the sw_flow_key reduc

[ovs-dev] [PATCH ovn] controller: Don't artificially limit group and meter IDs to 16bit.

OVS actually supports way more. Use the real numbers instead. To avoid preallocating huge bitmaps for the group/meter IDs, switch to id-pool instead (as suggested by Ilya). Reported-at: https://issues.redhat.com/browse/FDP-70 Suggested-by: Ilya Maximets Signed-off-by: Dumitru Ceara --- control

Re: [ovs-dev] [PATCH ovn 0/5] ci: ovn-kubernetes: Upstream sync and hardenning.

On 10/4/23 13:32, Ales Musil wrote: > On Wed, Oct 4, 2023 at 1:23 PM Dumitru Ceara wrote: >> >> Recently the ovn-kubernetes jobs failed often, mainly due to two >> reasons: >> - GitHub runners exhausting disk space >> - ovn-org/ovn job definitions being out of sync with >> ovn-org/ovn-kubernetes

[ovs-dev] [PATCH] Documentation: Add CVE-2022-40982, aka Downfall reference.

Added a reference to the DPDK documentation as a result of analyzing the OVS code for potential performance impacts due to the Downfall mitigation. Signed-off-by: Eelco Chaudron --- Documentation/topics/dpdk/bridge.rst |5 + 1 file changed, 5 insertions(+) diff --git a/Documentation/top

Re: [ovs-dev] [PATCH, v3] netdev: Sync'ed and cleaned {get, set}_config().

On 04.10.23 12:18, Ilya Maximets wrote: > On 10/2/23 16:44, David Marchand wrote: >> On Mon, Oct 2, 2023 at 1:52 PM Simon Horman wrote: >>> On Wed, Sep 27, 2023 at 03:24:07PM +0200, jm...@redhat.com wrote: From: Jakob Meng For better usability, the function pairs get_config() and >

[ovs-dev] [PATCH, v4] netdev: Sync'ed and cleaned {get, set}_config().

From: Jakob Meng For better usability, the function pairs get_config() and set_config() for each netdev should be symmetric: Options which are accepted by set_config() should be returned by get_config() and the latter should output valid options for set_config() only. This patch moves key-value

Re: [ovs-dev] [PATCH ovn 1/5] ci: Free up disk space in a more robust way.

On 10/4/23 13:39, 0-day Robot wrote: > Bleep bloop. Greetings Dumitru Ceara, 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 has non-spaces leading wh

Re: [ovs-dev] [PATCH ovn 1/5] ci: Free up disk space in a more robust way.

Bleep bloop. Greetings Dumitru Ceara, 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 has non-spaces leading whitespace WARNING: Line has trailing whitespace #63 FILE

Re: [ovs-dev] [PATCH ovn 0/5] ci: ovn-kubernetes: Upstream sync and hardenning.

On Wed, Oct 4, 2023 at 1:23 PM Dumitru Ceara wrote: > > Recently the ovn-kubernetes jobs failed often, mainly due to two > reasons: > - GitHub runners exhausting disk space > - ovn-org/ovn job definitions being out of sync with > ovn-org/ovn-kubernetes job definitions. > > This series tries to a

[ovs-dev] [PATCH ovn 5/5] ci: ovn-kubernetes: Synchronize CI jobs with ovn-org/ovn-kubernetes.

Update the set of jobs to match what ovn-kubernetes is currently running. We don't run everything though, just the conformance and control-plane tests. This is based on the following revision of ovn-kubernetes: https://github.com/ovn-org/ovn-kubernetes/tree/ac329f818509 Signed-off-by: Dumitru Ce

[ovs-dev] [PATCH ovn 4/5] ci: ovn-kubernetes: Reduce ovnkube pods log verbosity.

We're running out of space and these seem to be some of the large disk space consumers. Signed-off-by: Dumitru Ceara --- .github/workflows/ovn-kubernetes.yml |1 + 1 file changed, 1 insertion(+) diff --git a/.github/workflows/ovn-kubernetes.yml b/.github/workflows/ovn-kubernetes.yml index

[ovs-dev] [PATCH ovn 2/5] ci: Handle google-cloud-sdk -> google-cloud-cli package name change.

From: Ales Musil We remove this in the ovn-kubernetes runners to save space and the package changed name: https://github.com/actions/runner-images/commit/601b21b1a61d5d825cdb423e8a7a0f7985884581 Signed-off-by: Ales Musil --- .ci/linux-util.sh |2 +- 1 file changed, 1 insertion(+), 1 delet

[ovs-dev] [PATCH ovn 3/5] ci: Free up additional space for ovn-k jobs.

From: Ales Musil The ovn-kubernetes jobs require a lot of disk space, in order to avoid jobs being terminated due to low disk space [0] free up even more space. This is based on comments from runner-images project[1] [2]. As we do not use Android, .NET or Boost it should be perfectly fine to rem

[ovs-dev] [PATCH ovn 1/5] ci: Free up disk space in a more robust way.

Use apt patterns [0] to only try to remove real packages. Also unify the way we free up disk space for all workflows. [0] https://manpages.ubuntu.com/manpages/focal/man7/apt-patterns.7.html Signed-off-by: Dumitru Ceara --- .ci/linux-util.sh | 15 +++

[ovs-dev] [PATCH ovn 0/5] ci: ovn-kubernetes: Upstream sync and hardenning.

Recently the ovn-kubernetes jobs failed often, mainly due to two reasons: - GitHub runners exhausting disk space - ovn-org/ovn job definitions being out of sync with ovn-org/ovn-kubernetes job definitions. This series tries to address both problems. In an ideal world ovn-kubernetes would also u

Re: [ovs-dev] |fail| pw1842874 [ovs-dev, v3, branch-2.17, 2/2] conntrack: Remove nat_conn introducing key directionality.

Hi, Michael. Sorry to bother you again, but there seems to be still some issue with the testing on older branches. Could you, please, take another look? Best regards, Ilya Maximets. > Test-Label: intel-ovs-compilation > Test-Status: fail > http://patchwork.ozlabs.org/api/patches/1842874/ > >

Re: [ovs-dev] [PATCH, v3] netdev: Sync'ed and cleaned {get, set}_config().

On 9/27/23 15:24, jm...@redhat.com wrote: > From: Jakob Meng > > For better usability, the function pairs get_config() and > set_config() for each netdev should be symmetric: Options which are > accepted by set_config() should be returned by get_config() and the > latter should output valid optio

Re: [ovs-dev] [PATCH 3/3] netdev-dpdk: Document rx-steering status options.

On 04.10.23 10:32, jm...@redhat.com wrote: > From: Jakob Meng > > Fixes: fc06ea9a1883 ("netdev-dpdk: Add custom rx-steering configuration.") > --- > vswitchd/vswitch.xml | 14 ++ > 1 file changed, 14 insertions(+) > > diff --git a/vswitchd/vswitch.xml b/vswitchd/vswitch.xml > index 00

Re: [ovs-dev] [PATCH 2/3] netdev-dpdk: Fixed DPDK specific interface status options.

On 04.10.23 10:31, jm...@redhat.com wrote: > From: Jakob Meng > > The status options pci-vendor_id and pci-device_id for dpdk netdevs > have been replaced by bus_info. This patch updates the documentation > in vswitchd/vswitch.xml accordingly. > > Fixes: a77c7796f23a ("dpdk: Update to use v22.11.1

Re: [ovs-dev] [PATCH 1/3] netdev-dpdk: Document status options for VF MAC address.

On 04.10.23 10:26, jm...@redhat.com wrote: > From: Jakob Meng > > Fixes: f4336f504b17 ("netdev-dpdk: Add option to configure VF MAC address. ") > --- > vswitchd/vswitch.xml | 4 > 1 file changed, 4 insertions(+) > > diff --git a/vswitchd/vswitch.xml b/vswitchd/vswitch.xml > index cfcde34ff..

Re: [ovs-dev] [PATCH, v3] netdev: Sync'ed and cleaned {get, set}_config().

On 10/2/23 16:44, David Marchand wrote: > On Mon, Oct 2, 2023 at 1:52 PM Simon Horman wrote: >> >> On Wed, Sep 27, 2023 at 03:24:07PM +0200, jm...@redhat.com wrote: >>> From: Jakob Meng >>> >>> For better usability, the function pairs get_config() and >>> set_config() for each netdev should be sy

[ovs-dev] [PATCH v2 3/3] tests: Update some tests title prefix print.

Use test title prefix according to filename the test is in for tunnel.at and ofproto-dpif.at. Signed-off-by: Roi Dayan --- Notes: v2: - add prefix to bond cases. tests/ofproto-dpif.at | 10 +- tests/tunnel.at | 2 +- 2 files changed, 6 insertions(+), 6 deletions(-) diff

[ovs-dev] [PATCH v2 2/3] tests/tunnel.at: Add geneve options mirror test.

Test geneve options mirror flow doesn't add redundant mirror. Signed-off-by: Roi Dayan --- Notes: v2: - add dot in title. tests/tunnel.at | 29 + 1 file changed, 29 insertions(+) diff --git a/tests/tunnel.at b/tests/tunnel.at index ddeb66bc9fb7..c0b93211044

[ovs-dev] [PATCH v2 1/3] ofproto-dpif-upcall: Fix redundant mirror on geneve tunnel options.

The cited commit fixed missing mirror packets by reset mirror when packets are modified but setting geneve options was also treated as a modified packet but should be treated as a part of set_tunnel which doesn't reset mirror. Fixes: feed7f677505 ("ofproto-dpif-upcall: Mirror packets that are modi

[ovs-dev] [PATCH v2 0/3] Fix redundant mirror with tunnel options.

Hi, The first commit removing the redundant mirror when using tunnel options. The second is a test to verify it stays like this and doesn't break again. The third commit is just updating prefixes of tests to match the file their in. Thanks, Roi Roi Dayan (3): ofproto-dpif-upcall: Fix redundan

Re: [ovs-dev] [PATCH 3/3] netdev-dpdk: Document rx-steering status options.

Bleep bloop. Greetings Jakob Meng, 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 Jakob Meng needs to sign off. Lines checked: 40, Warnings: 0, Errors: 1 Please c

Re: [ovs-dev] [PATCH 1/1] tests: Update some tests title prefix print.

On 04/10/2023 12:46, Simon Horman wrote: > On Tue, Oct 03, 2023 at 05:03:20PM +0300, Roi Dayan via dev wrote: >> Use test title prefix according to filename the test is in >> for tunnel.at and ofproto-dpif.at. >> >> Signed-off-by: Roi Dayan >> --- >> tests/ofproto-dpif.at | 6 +++--- >> tests/t

Re: [ovs-dev] [PATCH 2/3] netdev-dpdk: Fixed DPDK specific interface status options.

Bleep bloop. Greetings Jakob Meng, 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 Jakob Meng needs to sign off. Lines checked: 40, Warnings: 0, Errors: 1 Please c

Re: [ovs-dev] [PATCH 1/3] netdev-dpdk: Document status options for VF MAC address.

Bleep bloop. Greetings Jakob Meng, 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 Jakob Meng needs to sign off. Lines checked: 30, Warnings: 0, Errors: 1 Please c

Re: [ovs-dev] [RFC PATCH 0/7] net: openvswitch: Reduce stack usage

On Mon Oct 2, 2023 at 9:54 PM AEST, Ilya Maximets wrote: > On 9/28/23 03:52, Nicholas Piggin wrote: > > On Wed Sep 27, 2023 at 6:36 PM AEST, Ilya Maximets wrote: > >> On 9/27/23 02:13, Nicholas Piggin wrote: > >>> Hi, > >>> > >>> We've got a report of a stack overflow on ppc64le with a 16kB kernel

Re: [ovs-dev] [PATCH 1/1] tests: Update some tests title prefix print.

On Tue, Oct 03, 2023 at 05:03:20PM +0300, Roi Dayan via dev wrote: > Use test title prefix according to filename the test is in > for tunnel.at and ofproto-dpif.at. > > Signed-off-by: Roi Dayan > --- > tests/ofproto-dpif.at | 6 +++--- > tests/tunnel.at | 2 +- > 2 files changed, 4 inserti

Re: [ovs-dev] [PATCH v5 1/1] dpif-netdev: Add per pmd sleep config.

Hello Kevin, On Fri, Sep 29, 2023 at 2:50 PM Kevin Traynor wrote: > > Extend 'pmd-sleep-max' so that individual PMD thread cores > may have a specified max sleep request value. > > Existing behaviour is maintained. > > Any PMD thread core without a value will use the global value > if set or defa

Re: [ovs-dev] [PATCH 3/3] netdev-dpdk: Document rx-steering status options.

On Wed, Oct 04, 2023 at 10:32:22AM +0200, jm...@redhat.com wrote: > From: Jakob Meng > > Fixes: fc06ea9a1883 ("netdev-dpdk: Add custom rx-steering configuration.") Acked-by: Simon Horman ___ dev mailing list d...@openvswitch.org https://mail.openvswi

Re: [ovs-dev] [PATCH 2/3] netdev-dpdk: Fixed DPDK specific interface status options.

On Wed, Oct 04, 2023 at 11:28:38AM +0200, Simon Horman wrote: > On Wed, Oct 04, 2023 at 10:31:25AM +0200, jm...@redhat.com wrote: > > From: Jakob Meng > > > > The status options pci-vendor_id and pci-device_id for dpdk netdevs > > have been replaced by bus_info. This patch updates the documentati

Re: [ovs-dev] [PATCH 2/3] netdev-dpdk: Fixed DPDK specific interface status options.

On Wed, Oct 04, 2023 at 10:31:25AM +0200, jm...@redhat.com wrote: > From: Jakob Meng > > The status options pci-vendor_id and pci-device_id for dpdk netdevs > have been replaced by bus_info. This patch updates the documentation > in vswitchd/vswitch.xml accordingly. > > Fixes: a77c7796f23a ("dpd

Re: [ovs-dev] [PATCH 1/3] netdev-dpdk: Document status options for VF MAC address.

On Wed, Oct 04, 2023 at 10:26:29AM +0200, jm...@redhat.com wrote: > From: Jakob Meng > > Fixes: f4336f504b17 ("netdev-dpdk: Add option to configure VF MAC address. ") Acked-by: Simon Horman ___ dev mailing list d...@openvswitch.org https://mail.openv

Re: [ovs-dev] [PATCH v2 2/3] dpctl, ovs-ofctl: Unify parsing of ct-flush arguments.

On Wed, Oct 4, 2023 at 8:45 AM Ales Musil wrote: > In order to make the command extensible unify the arguments > parsing into single function. This will be later on used > for the mark and labels arguments. > > Signed-off-by: Ales Musil > --- > include/openvswitch/ofp-ct.h | 5 ++-- > lib/dpct

Re: [ovs-dev] [PATCH ovn] northd: Incrementally process SB.Load_balancer updates.

On Wed, Oct 4, 2023 at 12:23 AM Dumitru Ceara wrote: > > Commit 9deb000536e0 ("northd: Remove potential duplicates in SB > Load_Balancer table.") added code to northd to ensure that SB duplicates > get removed. Quoting from that commit message: > > The Southbound Load_Balancer table (like the

Re: [ovs-dev] [PATCH] netdev: Fixed DPDK specific interface status options.

On 29.09.23 17:05, Kevin Traynor wrote: > On 28/09/2023 08:50, Simon Horman wrote: >> On Wed, Sep 27, 2023 at 01:46:19PM +0200, jm...@redhat.com wrote: >>> From: Jakob Meng >>> >>> The documentation in vswitchd/vswitch.xml for status columns has been >>> updated to reflect recent and not so recent

[ovs-dev] [PATCH 3/3] netdev-dpdk: Document rx-steering status options.

From: Jakob Meng Fixes: fc06ea9a1883 ("netdev-dpdk: Add custom rx-steering configuration.") --- vswitchd/vswitch.xml | 14 ++ 1 file changed, 14 insertions(+) diff --git a/vswitchd/vswitch.xml b/vswitchd/vswitch.xml index 006d1e6a4..1e2a1267d 100644 --- a/vswitchd/vswitch.xml +++ b/

[ovs-dev] [PATCH 2/3] netdev-dpdk: Fixed DPDK specific interface status options.

From: Jakob Meng The status options pci-vendor_id and pci-device_id for dpdk netdevs have been replaced by bus_info. This patch updates the documentation in vswitchd/vswitch.xml accordingly. Fixes: a77c7796f23a ("dpdk: Update to use v22.11.1.") --- vswitchd/vswitch.xml | 9 +++-- 1 file cha

[ovs-dev] [PATCH 1/3] netdev-dpdk: Document status options for VF MAC address.

From: Jakob Meng Fixes: f4336f504b17 ("netdev-dpdk: Add option to configure VF MAC address. ") --- vswitchd/vswitch.xml | 4 1 file changed, 4 insertions(+) diff --git a/vswitchd/vswitch.xml b/vswitchd/vswitch.xml index cfcde34ff..797fb05bf 100644 --- a/vswitchd/vswitch.xml +++ b/vswitchd/

Re: [ovs-dev] [PATCH 2/2] openflow: Allow CT flush to match on mark and labels.

On Tue, Oct 03, 2023 at 04:01:15PM +0200, Ilya Maximets wrote: > On 10/3/23 15:46, Ales Musil wrote: > > On Tue, Oct 3, 2023 at 3:40 PM Ilya Maximets > > wrote: > > On 10/3/23 07:56, Ales Musil wrote: ... > > > @@ -95,10 +139,27 @@ ofp_ct_match_format(struct ds

Re: [ovs-dev] [PATCH v2 2/2] conntrack: Use helpers from committed connections.

On Tue, Oct 03, 2023 at 04:10:47PM +0400, Viacheslav Galaktionov wrote: > On 9/29/23 17:02, Simon Horman wrote: > > On Mon, Sep 04, 2023 at 07:45:07PM +0400, Viacheslav Galaktionov via dev > > wrote: > > > Currently, if the user wants to track related connections, they have to > > > specify a help

Re: [ovs-dev] [RFC PATCH 4/7] net: openvswitch: ovs_vport_receive reduce stack usage

On Fri Sep 29, 2023 at 1:26 AM AEST, Aaron Conole wrote: > Nicholas Piggin writes: > > > Dynamically allocating the sw_flow_key reduces stack usage of > > ovs_vport_receive from 544 bytes to 64 bytes at the cost of > > another GFP_ATOMIC allocation in the receive path. > > > > XXX: is this a probl

Re: [ovs-dev] [RFC PATCH 4/7] net: openvswitch: ovs_vport_receive reduce stack usage

On Fri Sep 29, 2023 at 6:38 PM AEST, Eelco Chaudron wrote: > > > On 29 Sep 2023, at 9:00, Nicholas Piggin wrote: > > > On Fri Sep 29, 2023 at 1:26 AM AEST, Aaron Conole wrote: > >> Nicholas Piggin writes: > >> > >>> Dynamically allocating the sw_flow_key reduces stack usage of > >>> ovs_vport_rece