As reported by Ales, net/tap has broken support for checksum offloading.
Fixes have been sent to the DPDK side, but waiting for the fixes, simply
disable all conntrack related tests and some IPv6 tunnel tests.
Signed-off-by: David Marchand
---
tests/system-dpdk-macros.at | 9 +
1 file
polled by OVS.
Signed-off-by: David Marchand
---
tests/system-dpdk-macros.at| 85 +-
tests/system-dpdk-testsuite.at | 2 +
tests/system-dpdk.at | 3 --
3 files changed, 86 insertions(+), 4 deletions(-)
diff --git a/tests/system-dpdk-macros.at b/tests
Align system-dpdk existing helpers to other common OVS helpers so they
can accept some optional arguments.
Introduce a OVS_DPDK_STOP_VSWITCHD wrapper around OVS_VSWITCHD_STOP to
catch dpdk related logs in a centralised fashion.
Signed-off-by: David Marchand
---
tests/system-dpdk-macros.at
-by: David Marchand
---
tests/system-common-macros.at| 12
tests/system-offloads-traffic.at | 6 +++---
tests/system-traffic.at | 32
3 files changed, 31 insertions(+), 19 deletions(-)
diff --git a/tests/system-common-macros.at b/tests
on hugetlbfs presence.
Signed-off-by: David Marchand
Acked-by: Aaron Conole
---
tests/system-dpdk-macros.at | 7 +--
1 file changed, 1 insertion(+), 6 deletions(-)
diff --git a/tests/system-dpdk-macros.at b/tests/system-dpdk-macros.at
index 665ca5c2f0..01346fe6ec 100644
--- a/tests/system
for OVS main and PMD
threads).
Signed-off-by: David Marchand
Acked-by: Aaron Conole
---
Changes since v1:
- rebased after DPDK build has been moved out of linux-build.sh,
- restored running "normal" checks in the DPDK jobs,
---
.ci/dpdk-build.sh| 8 +---
Rather than copy/paste everywhere, introduce helpers to control
testpmd runs.
Rely on --stats-period (which outputs port stats every n seconds) so that
testpmd keeps running without expecting any user input.
Signed-off-by: David Marchand
Acked-by: Aaron Conole
---
Changes since v1:
- fixed
offloads.
Link:
https://patchwork.dpdk.org/project/dpdk/patch/20230818090351.2402519-1-david.march...@redhat.com/
Link:
https://patchwork.dpdk.org/project/dpdk/patch/20230822073244.3751885-1-david.march...@redhat.com/
Signed-off-by: David Marchand
---
lib/netdev-provider.h | 5 ++-
lib/netdev.c
: 5d11c47d3ebe ("userspace: Enable IP checksum offloading by default.")
Signed-off-by: David Marchand
---
Some note:
- I submitted a fix on the DPDK side for net/iavf. But to be on the safe
side, this OVS patch here is a proposal to avoid hitting bugs in some
other drivers,
---
lib/netdev-
On Tue, Aug 8, 2023 at 3:14 PM Ilya Maximets wrote:
>
> On 8/8/23 09:17, David Marchand wrote:
> > On Thu, Jul 13, 2023 at 11:02 AM Viacheslav Galaktionov via dev
> > wrote:
> >>
> >> The set_error function is now used regardless of whether experimental APIs
eering configuration.")
> Signed-off-by: Viacheslav Galaktionov
> Acked-by: Ivan Malov
Can we merge the dpdk-latest patch for checking experimental API
compilation in the master branch?
I am referring to:
https://github.com/openvswitch/ovs/commit/561e15fabd43613cb9c8a2d996c1da
On Mon, Aug 7, 2023 at 2:28 PM Ilya Maximets wrote:
>
> 13.2 was released in April and 13.1 images no longer able
> to update packages:
> pkg: repository FreeBSD contains packages for wrong OS version
>
> Fix that by updating to FreeBSD 13.2.
>
> Reported-by: Davi
at breaks the build.
>
> Fixes: 6cb2f5a630e3 ("github: Add GitHub Actions workflow.")
> Signed-off-by: Ilya Maximets
Reviewed-by: David Marchand
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
virtio features (and disabling TSO for this
vhost port).
Acked-by: Mike Pattrick
Acked-by: Simon Horman
Signed-off-by: David Marchand
---
Changelog since v5:
- fixed coding style,
Changelog since v4:
- I kept acks as the logic behind the state machine did not change much,
- fixed indent
On Wed, Jun 21, 2023 at 11:26 AM Kevin Traynor wrote:
>
> Signed-off-by: Kevin Traynor
Reviewed-by: David Marchand
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
On Wed, Jun 21, 2023 at 11:26 AM Kevin Traynor wrote:
>
> Add unit tests for new per pmd options of pmd-sleep-max.
>
> Signed-off-by: Kevin Traynor
Reviewed-by: David Marchand
--
David Marchand
___
dev mailing list
d...@openvswit
request values.
>
> Signed-off-by: Kevin Traynor
Reviewed-by: David Marchand
I have some nits below which can be fixed when applying.
But I am ok too if we go with the current patch.
> diff --git a/Documentation/topics/dpdk/pmd.rst
> b/Documentation/topics/dpdk/pmd.rst
> index 40
expanded when future
> additions are added.
>
> Signed-off-by: Kevin Traynor
Reviewed-by: David Marchand
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
On Wed, Jun 21, 2023 at 11:26 AM Kevin Traynor wrote:
>
> This is just cosmetic. There is no change to the tests.
>
> Signed-off-by: Kevin Traynor
Reviewed-by: David Marchand
--
David Marchand
___
dev mailing list
d...@openvswit
command can follow
> the 'pmd-sleep-xyz' pattern.
>
> Signed-off-by: Kevin Traynor
I don't have a strong opinion against this name change.
Having a clear prefix for this feature params and using this same
prefix for appctl commands (later in this series) seems a good thing
to me.
ng
is wrong on the TSO side.
With announcing that TSO is off for both sides of this vhost port, I
think we have more chances a user would search in OVS documentation
and find out about the "userspace-tso" status field.
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
virtio features (and disabling TSO for this
vhost port).
Acked-by: Mike Pattrick
Acked-by: Simon Horman
Signed-off-by: David Marchand
---
Changelog since v4:
- I kept acks as the logic behind the state machine did not change much,
- fixed indent of enumeration in documentation,
- used status
On Thu, Jun 29, 2023 at 9:43 PM Ilya Maximets wrote:
>
> On 6/20/23 13:26, David Marchand wrote:
> > At some point in OVS history, some virtio features were announced as
> > supported (ECN and UFO virtio features).
> >
> > The userspace TSO code, which has been
date the param name now for the reason Kevin gave in the
commitlog.
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
On Mon, Jun 26, 2023 at 4:05 PM David Marchand
wrote:
>
> On Mon, Jun 26, 2023 at 3:19 PM Ilya Maximets wrote:
> > > - if the UFO feature is "restored" in the master branch, OVS can't
> > > expose CSUM if the guest negotiated UFO.
> >
> > Yep. I
e for new/restarted VMs.
> >>
> >> Does that make sense?
> >
> > I think you covered all cases.
I have been thinking about it again.
From a guest pov, is it important to support checksum offloading if
TSO is not working?
I am not sure anymore what was the reason to have only a
On Mon, Jun 26, 2023 at 11:37 AM David Marchand
wrote:
> > Currently possible configurations:
> >
> > 1. CSUM + TSO + ___ + ___ OVS 3.1 with userspace-tso=true
> > 2. + TSO + ECN + ___ OVS 3.1 with userspace-tso=false
> > 3. + ___ + ECN + UFO O
__ + ___
> y. CSUM + ___ + ___ + ___
>
>
> Logic can be:
>
> 1. Try 'a/x' if userspace-tso=true
> 2. Try 'y' if userspace-tso=false
> 3. Try 'c' if above failed
>
> As far as I understand we can advertise new features, but we can't remove
> already acked ones. So, the configuration 'c' is a catch-all for broken
> cases. 'y' will only be possible for new/restarted VMs.
>
> Does that make sense?
I think you covered all cases.
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
On Fri, Jun 23, 2023 at 5:14 PM Ilya Maximets wrote:
> On 6/23/23 15:16, David Marchand wrote:
> > On Wed, Jun 21, 2023 at 12:13 AM Ilya Maximets wrote:
> >>
> >> On 6/14/23 21:03, Mike Pattrick wrote:
> >>> The netdev receiving packets is supposed to provi
_ANY_LAYOUT
VIRTIO_RING_F_INDIRECT_DESC
VIRTIO_RING_F_EVENT_IDX
! VHOST_USER_F_PROTOCOL_FEATURES
VIRTIO_F_VERSION_1
VIRTIO_F_RING_PACKED
VIRTIO_F_IN_ORDER
And.. so qemu sets VHOST_USER_SET_FEATURES:
$ ./features.sh 6176171906
VIRTIO_NET_F_GUEST_CSUM
VIRTIO_NET_F_GUEST_TSO4
VIRTIO_NET_F_GUEST_TSO6
VIRTIO_NET_F_GUEST_ECN
VIRTIO_NET_F_GUEST_UFO
VIRTIO_NET_F_HOST_ECN
VIRTIO_NET_F_HOST_UFO
VIRTIO_NET_F_MRG_RXBUF
VIRTIO_NET_F_GUEST_ANNOUNCE
VIRTIO_RING_F_INDIRECT_DESC
VIRTIO_RING_F_EVENT_IDX
! VHOST_USER_F_PROTOCOL_FEATURES
VIRTIO_F_VERSION_1
As a conclusion, we may have to handle both incorrect setting of UFO
and ECN :-(.
We probably missed this issue during validation as people are not
testing upgrades from 2.11 to current master.
Did I miss any other concern of yours?
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
virtio features (and disabling TSO for this
vhost port).
Signed-off-by: David Marchand
Acked-by: Mike Pattrick
---
Note: resending this v4 for CI (I added Mike ack).
Changelog since v3:
- updated documentation now that the interface offloads status is reported
in ovsdb,
- fixed one coding style
(void *)_aux);
> +unixctl_command_register("dpif-netdev/pmd-sleep-show", "[-pmd core]
> [dp]",
> + 0, 3, dpif_netdev_pmd_info,
> + (void *)_aux);
> unixctl_command_register("dp
_UNTIL([tail ovs-vswitchd.log | grep "PMD load based sleeps are
> disabled."])
> +CHECK_DP_SLEEP_MAX([0], [disabled], [])
>
> dnl Check low value max sleep
> get_log_next_line_num
Not directly related to this patch.
I did not realise before when those tests were added, b
virtio features (and disabling TSO for this
vhost port).
Signed-off-by: David Marchand
---
Note: this series depends on Mike series and won't apply cleanly
without it.
Changelog since v3:
- updated documentation now that the interface offloads status is reported
in ovsdb,
- fixed one coding style
cenario deleting vxlan ports, flushing flows
manually and I saw nothing broken.
Reviewed-by: David Marchand
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
for details.
> ovs-vsctl: The default log directory is "/var/log/openvswitch".
Ok, got it thanks.
It seems a bit strange to remove and re-add the PF port, but cleaning
up those flows makes sense on the principle.
>
> I can rebase and rephrase the commit msg.
Maybe rephrase yes, otherwise the fix works and lgtm.
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
Fixes: b366fa2f4947 ("dpif-netdev: Call cpuid for x86 isa availability.")
Signed-off-by: David Marchand
---
lib/cpu.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/lib/cpu.c b/lib/cpu.c
index 0292f715ec..fbbea40053 100644
--- a/lib/cpu.c
+++ b/lib/cpu.c
@@ -
On Wed, May 31, 2023 at 3:22 PM David Marchand
wrote:
>
> Caught while reviewing code.
>
> Fixes: aca2f8a8a6b6 ("netdev-offload-dpdk: Implement HW miss packet recover
> for vport.")
> Fixes: b6cabb8f8f80 ("netdev: Split up netdev offloading to separate module.&q
Caught while reviewing code.
Fixes: aca2f8a8a6b6 ("netdev-offload-dpdk: Implement HW miss packet recover for
vport.")
Fixes: b6cabb8f8f80 ("netdev: Split up netdev offloading to separate module.")
Signed-off-by: David Marchand
---
lib/netdev-offload-dpdk.c | 2 +-
lib/netde
you send a v2, feel free to add:
Reviewed-by: David Marchand
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
sages.")
Signed-off-by: David Marchand
---
lib/netdev-offload-dpdk.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/netdev-offload-dpdk.c b/lib/netdev-offload-dpdk.c
index 7c91f8a799..14bc877719 100644
--- a/lib/netdev-offload-dpdk.c
+++ b/lib/netdev-offload-dpdk.c
@
On Fri, May 26, 2023 at 4:35 PM David Marchand
wrote:
>
> Hello Eli,
>
> On Mon, Sep 5, 2022 at 4:46 PM Eli Britstein via dev
> wrote:
> >
> > Vport's offloads are done on the tracked orig-in-port, but the flow itself
> > is associated in the vport's map.
> &
49 ("netdev-offload-dpdk: Use per-netdev offload metadata.")
> Reported-by: 15895987278
> Signed-off-by: Eli Britstein
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
; value and log.
>
> Reported-at: https://bugzilla.redhat.com/2119876
> Signed-off-by: Kevin Traynor
Lgtm.
Reviewed-by: David Marchand
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
aximets
> Signed-off-by: Kevin Traynor
Reviewed-by: David Marchand
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
EFAULT_RXQ_SIZE,
> ->requested_rxq_size);
> -dpdk_process_queue_size(netdev, args, "n_txq_desc",
> -NIC_PORT_DEFAULT_TXQ_SIZE,
> ->requested_txq_size);
> -
> new_dev
DPDK does not depend on how we will link OVS against it, so we
can use a single cache entry regardless of DPDK_SHARED option,
Reviewed-by: Simon Horman
Signed-off-by: David Marchand
---
Changes since v3:
- fixed git log trace,
- set DPDK_GIT explicitly in yml,
Changes since v2:
- removed
dpdk-ci-signature
> > +grep -rwE 'DPDK_GIT|DPDK_VER' .github/ >> dpdk-ci-signature
> > +if [ "${DPDK_VER##refs/*/}" != "${DPDK_VER}" ]; then
> > +git ls-remote --heads $DPDK_GIT $DPDK_VER >> dpdk-ci-signature
>
>
On Fri, Apr 28, 2023 at 2:26 PM David Marchand
wrote:
>
> On Fri, Apr 28, 2023 at 11:36 AM David Marchand
> wrote:
> > > > +- name: create ci signature file for the dpdk cache key
> > > > + # This will collect most of DPDK related lines, s
DPDK does not depend on how we will link OVS against it, so we
can use a single cache entry regardless of DPDK_SHARED option,
Reviewed-by: Simon Horman
Signed-off-by: David Marchand
---
Changes since v2:
- removed unneeded checks when the cache is hit,
- exposed the cache key computed
On Fri, Apr 28, 2023 at 11:36 AM David Marchand
wrote:
> > > +- name: create ci signature file for the dpdk cache key
> > > + # This will collect most of DPDK related lines, so hash will be
> > > different
> > > + # if something changed in
DPDK including
> > DPDK_VER.
> > + # This also allows us to use cache from any branch as long as version
> > + # and a way we're building DPDK stays the same.
> > + run: |
> > +grep -irE 'RTE_|DPDK|meson|ninja' .ci/dpdk-* > dpdk-ci-signature
>
onfigure vswitchd/bridge.c:885
> #16 0x41f1a9 in bridge_run vswitchd/bridge.c:3313
> #17 0x42d4fb in main vswitchd/ovs-vswitchd.c:132
> #18 0x7fe09cc03c86 in __libc_start_main (/usr/lib64/libc.so.6+0x25c86)
>
> Fixes: b841e3cd4a28 ("dpif-netlink: Fix feature negotiation
; value and log.
>
> Reported-at: https://bugzilla.redhat.com/2119876
> Signed-off-by: Kevin Traynor
Thanks for the added log.
Overall, this looks good to me.
I have a comment on how the limits are taken into account (see below).
But in any case this patch is ok as is for me:
Reviewed-by: D
nitially when enabling ASan:
https://mail.openvswitch.org/pipermail/ovs-dev/2021-February/380467.html
More recently, Eelco mentionned it:
https://mail.openvswitch.org/pipermail/ovs-dev/2022-September/397476.html
I also see it every once in a while in GHA.
--
David Marchand
_
specific call to dpif_netlink_dp_transact() (line 398) in
dpif_netlink_open(), the 'dp' content is not considered in the branch
when no error returned (starting line 430).
Besides, 'dp' / 'buf' are overwritten later in this same branch, by
sending a new netlink request (line 437).
So if this code is not
xes tag for your patch.
I just sent a new patch, and marked this current one as Rejected.
Thanks Simon and sorry for the confusion.
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
Caught during some code review.
SUPPORT_TC_INGRESS_PPS has been replaced with CHECK_TC_INGRESS_PPS().
Fixes: 5f0fdf5e2c2e ("test: Move check for tc ingress pps support to test
script.")
Signed-off-by: David Marchand
---
tests/system-offloads-traffic.at | 2 +-
1 file changed, 1 inser
_request.user_features |= OVS_DP_F_VPORT_PIDS;
dp_request.user_features |= OVS_DP_F_UNSUPPORTED;
-error = dpif_netlink_dp_transact(_request, , );
+error = dpif_netlink_dp_transact(_request, NULL, NULL);
if (error) {
/* The Open vSwitch kernel module has two modes for dispatching
On Tue, Apr 18, 2023 at 10:20 AM Simon Horman wrote:
>
> On Mon, Apr 17, 2023 at 06:28:13PM +0200, David Marchand wrote:
> > Caught during some code review.
> > The incriminated commit had put an unneeded check on tc ingress support
> > for the meter offlo
60b89a309d ("dpif-netlink: Offloading meter to tc police action")
Signed-off-by: David Marchand
---
tests/system-offloads-traffic.at | 1 -
1 file changed, 1 deletion(-)
diff --git a/tests/system-offloads-traffic.at b/tests/system-offloads-traffic.at
index da18597cd8..df43caa9e5 10064
DPDK does not depend on how we will link OVS against it, so we
can use a single cache entry regardless of DPDK_SHARED option,
Signed-off-by: David Marchand
---
Changes since v1:
- filtered dpdk build job dependencies: it only needs gcc, ninja and
libnuma-dev,
- removed matrix configuration
Hello Ilya,
On Tue, Apr 4, 2023 at 8:01 PM Ilya Maximets wrote:
>
> On 3/30/23 11:24, David Marchand wrote:
> > Let's separate DPDK compilation from the rest of OVS build:
> > - this avoids multiple jobs building DPDK in parallel, which especially
> > affects builds i
s.
> We may switch to quay.io if that will ever become a problem
> in the future.
>
> Signed-off-by: Ilya Maximets
Reviewed-by: David Marchand
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
er.
I would go with ubuntu-latest.
> +container: fedora:37
> +timeout-minutes: 30
> +
> +strategy:
> + fail-fast: false
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
On Thu, Mar 30, 2023 at 11:24 AM David Marchand
wrote:
>
> Let's separate DPDK compilation from the rest of OVS build:
> - this avoids multiple jobs building DPDK in parallel, which especially
> affects builds in the dpdk-latest branch,
Err, this patch actually breaks dpdk-
DPDK does not depend on how we will link OVS against it, so we
can use a single cache entry regardless of DPDK_SHARED option,
Signed-off-by: David Marchand
---
.ci/dpdk-build.sh| 68
.ci/dpdk-prepare.sh | 11 +
.ci/linux
if (retval) {
> diff --git a/vswitchd/ovs-vswitchd.8.in b/vswitchd/ovs-vswitchd.8.in
> index 9569265fcb..a6a4a24606 100644
> --- a/vswitchd/ovs-vswitchd.8.in
> +++ b/vswitchd/ovs-vswitchd.8.in
> @@ -81,6 +81,14 @@ unavailable or unsuccessful.
>
On Mon, Mar 6, 2023 at 5:05 PM David Marchand wrote:
> > diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
> > index fb0dd43f7..e901f857e 100644
> > --- a/lib/netdev-dpdk.c
> > +++ b/lib/netdev-dpdk.c
> > @@ -1008,4 +1008,13 @@ dpdk_watchdog(void *dummy OVS_UNUS
ntel xl710, which reserves some of them for
> --
> 2.39.1
>
> ___
> dev mailing list
> d...@openvswitch.org
> https://mail.openvswitch.org/mailman/listinfo/ovs-dev
>
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
fg->filter;
I am not familiar with this code.
I prefer to ask.
Is it safe to only reference the ovsrec content?
I think it is in this case, since mirror_set will convert this
information into a flow / mask.
> +
> /* Get port selection. */
> if (cfg->select_all) {
>
rect in fdset_try_del ?
> >
> >CC: David and Maxime.
This could be a different issue, but there was a change for this area
of the vhost library in v21.11:
451dc0fad83d ("vhost: fix crash on port deletion")
I don't see it backported to 19.11 stable release, but it got to
v20.11.4 (used in 2.15 and 2.16 ovs branches).
I would suggest backporting it to your 19.11 dpdk and retest with ovs 2.14.
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
error: Expected ) in function call
[...]
Besides, the list of headers by rte_memcpy.h is now out of sync with DPDK.
OVS takes care to include the right headers in its sources.
Simply make this header self-sufficient.
1: https://gcc.gnu.org/git/?p=gcc.git;a=commit;h=a68412117fa4
Signed-of
On Fri, Feb 3, 2023 at 4:13 PM Ilya Maximets wrote:
>
> On 2/2/23 10:21, David Marchand wrote:
> > On Wed, Feb 1, 2023 at 5:54 PM David Marchand
> > wrote:
> >>
> >> rte_vect.h pulls some AVX512 instrinsics headers added in GCC 12 [1]
> >> trigger
On Fri, Jan 27, 2023 at 12:15 PM Ilya Maximets wrote:
>
> On 1/27/23 11:29, David Marchand wrote:
> > On Thu, Jan 26, 2023 at 11:30 PM Aaron Conole wrote:
> >>
> >> David Marchand writes:
> >>
> >>> Let's enhance our coverage in the CI and ru
On Fri, Jan 27, 2023 at 12:13 PM Ilya Maximets wrote:
>
> On 1/23/23 13:14, David Marchand wrote:
> > Let's enhance our coverage in the CI and run DPDK system tests.
> >
> > A few DPDK drivers are enabled in DPDK compilation.
> >
> > Put DPDK build in $P
On Wed, Feb 1, 2023 at 5:54 PM David Marchand wrote:
>
> rte_vect.h pulls some AVX512 instrinsics headers added in GCC 12 [1]
> trigger a lot of warnings:
>
> libtool: compile: env "REAL_CC=ccache gcc" "CHECK=sparse -Wsparse-error
> -I ../include/sparse
h both versions of a header.
> Just defining all the functions we actually use in OVS instead and
> not including the original header.
>
> Fixes: e8568993e062 ("netdev-afxdp: NUMA-aware memory allocation for XSK
> related memory.")
> Signed-off-by: Ilya Maximet
re should be no side effect in removing those headers inclusions from
the rte_memcpy.h fake header.
1: https://gcc.gnu.org/git/?p=gcc.git;a=commit;h=a68412117fa4
Signed-off-by: David Marchand
---
include/sparse/rte_memcpy.h | 6 --
1 file changed, 6 deletions(-)
diff --git a/include/s
ng - offloads enabled"
>
> Fixes: 5660b89a309d ("dpif-netlink: Offloading meter to tc police action")
> Reported-by: David Marchand
> Reviewed-by: Louis Peens
> Signed-off-by: Simon Horman
Reviewed-by: David Marchand
--
David Marchand
t.")
> Reported-by: David Marchand
> Reviewed-by: Louis Peens
> Signed-off-by: Simon Horman
Reviewed-by: David Marchand
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
On Sun, Jan 29, 2023 at 11:34 AM Simon Horman wrote:
> > >>>> Probably not important, but:
> > >>>> Fixes: b020a416e24c ("System Tests: Enhance NAT tests.")
> > >>>>> Signed-off-by: Ilya Maximets
> > >>>>
On Thu, Jan 26, 2023 at 11:30 PM Aaron Conole wrote:
>
> David Marchand writes:
>
> > Let's enhance our coverage in the CI and run DPDK system tests.
> >
> > A few DPDK drivers are enabled in DPDK compilation.
> >
> > Put DPDK build in $PATH for dpdk-testp
On Mon, Jan 23, 2023 at 1:14 PM David Marchand
wrote:
> +# OVS_DPDK_START_TESTPMD()
> +#
> +# Start dpdk-testpmd in background.
> +#
> +m4_define([OVS_DPDK_START_TESTPMD],
> + [AT_CHECK([lscpu], [], [stdout])
> + AT_CHECK([cat stdout | grep "NUMA node(s)&qu
On Mon, Jan 23, 2023 at 3:50 PM Simon Horman wrote:
>
> On Mon, Jan 23, 2023 at 03:29:50PM +0100, David Marchand wrote:
> > On Mon, Jan 23, 2023 at 3:05 PM Ilya Maximets wrote:
> > >
> > > Test fails is 'nc' is not available, it should be skipped instead.
> &
On Mon, Jan 23, 2023 at 3:05 PM Ilya Maximets wrote:
>
> Test fails is 'nc' is not available, it should be skipped instead.
>
Probably not important, but:
Fixes: b020a416e24c ("System Tests: Enhance NAT tests.")
> Signed-off-by: Ilya Maximets
Reviewed-by: David
).
Signed-off-by: David Marchand
---
.ci/linux-build.sh | 22 ++
.github/workflows/build-and-test.yml | 7 ---
tests/system-dpdk-macros.at | 2 +-
3 files changed, 23 insertions(+), 8 deletions(-)
diff --git a/.ci/linux-build.sh b/.ci/linux
on hugetlbfs presence.
Signed-off-by: David Marchand
---
tests/system-dpdk-macros.at | 7 +--
1 file changed, 1 insertion(+), 6 deletions(-)
diff --git a/tests/system-dpdk-macros.at b/tests/system-dpdk-macros.at
index 08117d3bf9..9f13f8ec20 100644
--- a/tests/system-dpdk-macros.at
+++ b
Rather than copy/paste everywhere, introduce helpers to control
testpmd runs.
Rely on --stats-period (which outputs port stats every n seconds) so that
testpmd keeps running without expecting any user input.
Signed-off-by: David Marchand
---
tests/system-dpdk-macros.at | 34
tests
=algorithms.Blowfish,
/usr/lib/python3.11/site-packages/scapy/layers/ipsec.py:485:
CryptographyDeprecationWarning: CAST5 has been deprecated
cipher=algorithms.CAST5,
Signed-off-by: Robin Jarry
Signed-off-by: David Marchand
---
tests/mfex_fuzzy.py | 8
1 file changed, 8 insertions
erfaces that might be sensitive to large sleep times.
>
> Signed-off-by: Kevin Traynor
Reviewed-by: David Marchand
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
ch.org/pipermail/ovs-dev/2023-January/401101.html
Fixes: 3b29286db1c5 ("netdev-dpdk: Add per virtqueue statistics.")
Signed-off-by: David Marchand
---
Changes since v1:
- moved handling of events to a dedicated thread,
- added a warning if a fair number of updates is pending,
---
lib/net
ng updates.
Now that I went with the dedicated thread approach, all I see is an
increase of 1-2M, maximum.
I think we can avoid going with your suggestion.
WDYT?
>
> We're entering a gross over-engineering territory, but the solution
> might be following:
>
> RCU-list with nodes with following elements:
> - ifname
> - qid
> - atomic_bool enabled
> - atomic_bool seen
> - timestamp
--
David Marchand
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
On Fri, Jan 13, 2023 at 1:58 PM Ilya Maximets wrote:
>
> On 1/12/23 21:55, David Marchand wrote:
> > As Ilya reported, we have a ABBA deadlock between DPDK vq->access_lock
> > and OVS dev->mutex when OVS main thread refreshes statistics, while a
> > vring state c
}
> > +mpsc_queue_release(_state_change_queue);
>
> Maybe we should limit the number of iteration at each run, to give time
> for the OVS main thread to handle other work in case the queue gets
> overloaded (intentionally or not).
I agree that this is somethi
g from the vhost-events DPDK thread, back to the OVS main thread
using a lockless queue.
Reported-at:
https://mail.openvswitch.org/pipermail/ovs-dev/2023-January/401101.html
Fixes: 3b29286db1c5 ("netdev-dpdk: Add per virtqueue statistics.")
Signed-off-by: David Marchand
---
lib/ne
On Thu, Jan 12, 2023 at 1:26 PM Ilya Maximets wrote:
>
> On 1/6/23 16:58, David Marchand wrote:
> > The DPDK vhost-user library maintains more granular per queue stats
> > which can replace what OVS was providing for vhost-user ports.
> >
> > The benefits for O
uf_pool_bulk() - that tests with
> different mempools.
>
>
> I vaguely remember that there was an API to bulk free mbufs that was
> suitable only when all mbufs are from the same pool. And the comment
> that I'm removing here in the code also suggests that. But I don't
> remember wh
On Wed, Jan 11, 2023 at 8:34 PM Ilya Maximets wrote:
>
> On 1/9/23 17:55, David Marchand wrote:
> > This archive website disappeared.
> > On the other hand, the link to an obsolete dpif-provider man page
> > probably did not provide much info and we can simply ment
This archive website disappeared.
On the other hand, the link to an obsolete dpif-provider man page
probably did not provide much info and we can simply mention the current
file.
Signed-off-by: David Marchand
---
Changes since v1:
- renumbered links,
---
Documentation/topics/windows.rst | 29
201 - 300 of 990 matches
Mail list logo
