Re: release candidate 23.03-rc3

[David Marchand <david.marchand@redhat.com>]

On Mon, Mar 20, 2023 at 5:29 PM Thomas Monjalon <thomas@monjalon.net> wrote:
>
> A new DPDK release candidate is ready for testing:
>         https://git.dpdk.org/dpdk/tag/?id=v23.03-rc3
>
> There are 110 new patches in this snapshot.
>
> Release notes:
>         https://doc.dpdk.org/guides/rel_notes/release_23_03.html
>
> Few major additions arrived in this -rc3:
>         - test application for machine learning inference
>         - DTS hello world
> The test framework DTS is being improved and migrated into the mainline.
> Please join the DTS effort for contributing, reviewing or testing.
>
> As usual, you can report any issue on https://bugs.dpdk.org
>
> DPDK 23.03-rc4 could be out at the end of this week.

A regression has been raised against v23.03-rc3 by Red Hat QE (see
tests with FAILED status in the report below).

This regression has been tracked in bz 1196 and fixed with
9e3223adfab7 ("vhost: fix deadlock with no multiqueue").


Test environment:

kernel 5.14
qemu 6.2
dpdk: git://dpdk.org/dpdk

# git log -1
                          commit 9e3223adfab71bc88e9ba67c520b87c529d34a35
                          Author: David Marchand <david.marchand@redhat.com>
                          Date:   Thu Mar 23 15:44:33 2023 +0100
                              vhost: fix deadlock with no multiqueue

                              This deadlock happens when a guest, that
had virtio ports with multi
                              queues configured, does not announce the
multi q feature in
                              SET_FEATURES.
                              In such a situation, all vq locks are
already taken before calling
                              free_vq(), which itself takes the lock.
                              As mentioned in the code, in this
situation, the virtio device is not
                              running yet and no datapath thread is
using the vq. So we can
                              release the lock before calling free_vq().

                              Bugzilla ID: 1196
                              Fixes: 4b02c2673757 ("vhost: annotate
async accesses")

                              Signed-off-by: David Marchand
<david.marchand@redhat.com>
                              Reviewed-by: Maxime Coquelin
<maxime.coquelin@redhat.com>

NICs: X540-AT2 NIC(ixgbe, 10G)

Test result:

Guest with device assignment(PF) throughput testing(1G hugepage size): PASS
Guest with device assignment(PF) throughput testing(2M hugepage size) : PASS
Guest with device assignment(VF) throughput testing: PASS
Guest with vhost-user 2Q throughput testing: PASS
Vhost-user reconnect with dpdk-client, qemu-server: qemu reconnect: PASS
Vhost-user reconnect with dpdk-client, qemu-server: ovs reconnect: PASS
Guest with ovs+dpdk+vhost-user 1Q live migration testing: PASS
Guest with ovs+dpdk+vhost-user 1Q live migration testing (2M): PASS
Guest with ovs+dpdk+vhost-user 2Q live migration testing: PASS
Guest with ovs+dpdk+vhost-user 4Q live migration testing: PASS
Host PF + DPDK testing: PASS
Host VF + DPDK testing: PASS
PVP 1Q live migration testing: PASS
PVP (host dpdk testpmd as vswitch) 1Q: throughput testing: PASS
PVP vhost-user 2Q throughput testing: FAILED  -->  PASS
PVP vhost-user 1Q - cross numa node throughput testing:  FAILED  -->  PASS
PVP 1Q cross numa node live migration testing: FAILED  -->  PASS
PVP 4Q  reconnect with dpdk-client, qemu-server dpdk-testpmd
reconnect: FAILED  -->  PASS


-- 
David Marchand