dpdk: Update to use DPDK v21.11.

This commit adds support for DPDK v21.11, it includes the following
changes.

1. ci: Install python elftools for DPDK 21.02.
2. ci: Update meson requirement for DPDK 21.05.
3. netdev-dpdk: Fix build with 21.05.
4. ci: Compile DPDK in non developer mode.

   http://patchwork.ozlabs.org/project/openvswitch/list/?series=242480&state=*

5. netdev-dpdk: Remove access to DPDK internals.
6. netdev-dpdk: Remove unused attribute from rte_flow rule.
7. netdev-dpdk: Fix mbuf macros namespace with 21.11-rc1.
8. netdev-dpdk: Fix vhost namespace with 21.11-rc2.

   http://patchwork.ozlabs.org/project/openvswitch/list/?series=271159&state=*

In addition documentation and DPDK unit tests were also updated in this
commit for use with DPDK v21.11.

For credit all authors of the original commits to 'dpdk-latest' with the above
changes have been added as co-authors for this commit.

Signed-off-by: David Marchand <david.marchand@redhat.com>
Co-authored-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Tested-by: Emma Finn <emma.finn"intel.com>
Tested-by: Seamus Ryan <seamus.ryan@intel.com>
Acked-by: Kevin Traynor <ktraynor@redhat.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>

.ci/linux-build.sh

@@ -159,6 +159,10 @@ function install_dpdk()
     # Disable building DPDK unit tests. Not needed for OVS build or tests.
     DPDK_OPTS="$DPDK_OPTS -Dtests=false"
 
+    # Disable DPDK developer mode, this results in less build checks and less
+    # meson verbose outputs.
+    DPDK_OPTS="$DPDK_OPTS -Ddeveloper_mode=disabled"
+
     # Install DPDK using prefix.
     DPDK_OPTS="$DPDK_OPTS --prefix=$(pwd)/build"
 
@@ -216,7 +220,7 @@ fi
 
 if [ "$DPDK" ] || [ "$DPDK_SHARED" ]; then
     if [ -z "$DPDK_VER" ]; then
-        DPDK_VER="20.11.1"
+        DPDK_VER="21.11"
     fi
     install_dpdk $DPDK_VER
     if [ "$CC" = "clang" ]; then

.ci/linux-prepare.sh

@@ -21,8 +21,8 @@ make -j4 HAVE_LLVM= HAVE_SQLITE= install
 cd ..
 
 pip3 install --disable-pip-version-check --user \
-    flake8 hacking sphinx wheel setuptools
+    flake8 hacking sphinx wheel setuptools pyelftools
-pip3 install --user  'meson==0.47.1'
+pip3 install --user  'meson==0.49.2'
 
 if [ "$M32" ]; then
     # Installing 32-bit libraries.

Documentation/faq/releases.rst

@@ -225,7 +225,7 @@ Q: Are all the DPDK releases that OVS versions work with maintained?
     The latest information about DPDK stable and LTS releases can be found
     at `DPDK stable`_.
 
-.. _DPDK stable: http://doc.dpdk.org/guides-20.11/contributing/stable.html
+.. _DPDK stable: http://doc.dpdk.org/guides-21.11/contributing/stable.html
 
 Q: I get an error like this when I configure Open vSwitch:
 

Documentation/intro/install/dpdk.rst

@@ -42,7 +42,7 @@ Build requirements
 In addition to the requirements described in :doc:`general`, building Open
 vSwitch with DPDK will require the following:
 
-- DPDK 20.11.1
+- DPDK 21.11
 
 - A `DPDK supported NIC`_
 
@@ -59,8 +59,8 @@ vSwitch with DPDK will require the following:
 
 Detailed system requirements can be found at `DPDK requirements`_.
 
-.. _DPDK supported NIC: https://doc.dpdk.org/guides-20.11/nics/index.html
+.. _DPDK supported NIC: https://doc.dpdk.org/guides-21.11/nics/index.html
-.. _DPDK requirements: https://doc.dpdk.org/guides-20.11/linux_gsg/sys_reqs.html
+.. _DPDK requirements: https://doc.dpdk.org/guides-21.11/linux_gsg/sys_reqs.html
 
 .. _dpdk-install:
 
@@ -73,9 +73,9 @@ Install DPDK
 #. Download the `DPDK sources`_, extract the file and set ``DPDK_DIR``::
 
        $ cd /usr/src/
-       $ wget https://fast.dpdk.org/rel/dpdk-20.11.1.tar.xz
+       $ wget https://fast.dpdk.org/rel/dpdk-21.11.tar.xz
-       $ tar xf dpdk-20.11.1.tar.xz
+       $ tar xf dpdk-21.11.tar.xz
-       $ export DPDK_DIR=/usr/src/dpdk-stable-20.11.1
+       $ export DPDK_DIR=/usr/src/dpdk-21.11
        $ cd $DPDK_DIR
 
 #. Configure and install DPDK using Meson
@@ -121,7 +121,7 @@ Install DPDK
 
 .. _DPDK sources: http://dpdk.org/rel
 .. _DPDK documentation:
-   https://doc.dpdk.org/guides-20.11/linux_gsg/build_dpdk.html
+   https://doc.dpdk.org/guides-21.11/linux_gsg/build_dpdk.html
 
 Install OVS
 ~~~~~~~~~~~
@@ -722,7 +722,7 @@ Limitations
   release notes`_.
 
 .. _DPDK release notes:
-   https://doc.dpdk.org/guides-20.11/rel_notes/release_20_11.html
+   https://doc.dpdk.org/guides-21.11/rel_notes/release_21_11.html
 
 - Upper bound MTU: DPDK device drivers differ in how the L2 frame for a
   given MTU value is calculated e.g. i40e driver includes 2 x vlan headers in

Documentation/topics/dpdk/phy.rst

@@ -117,7 +117,7 @@ tool::
 
 For more information, refer to the `DPDK documentation <dpdk-drivers>`__.
 
-.. _dpdk-drivers: https://doc.dpdk.org/guides-20.11/linux_gsg/linux_drivers.html
+.. _dpdk-drivers: https://doc.dpdk.org/guides-21.11/linux_gsg/linux_drivers.html
 
 .. _dpdk-phy-multiqueue:
 
@@ -235,7 +235,7 @@ To hotplug a port with igb_uio in this case, DPDK must be configured to use
 physical addressing for IOVA mode. For more information regarding IOVA modes
 in DPDK please refer to the `DPDK IOVA Mode Detection`__.
 
-__ https://doc.dpdk.org/guides-20.11/prog_guide/env_abstraction_layer.html#iova-mode-detection
+__ https://doc.dpdk.org/guides-21.11/prog_guide/env_abstraction_layer.html#iova-mode-detection
 
 To configure OVS DPDK to use physical addressing for IOVA::
 
@@ -267,7 +267,7 @@ Representors are multi devices created on top of one PF.
 
 For more information, refer to the `DPDK documentation`__.
 
-__ https://doc.dpdk.org/guides-20.11/prog_guide/switch_representation.html
+__ https://doc.dpdk.org/guides-21.11/prog_guide/switch_representation.html
 
 Prior to port representors there was a one-to-one relationship between the PF
 and the eth device. With port representors the relationship becomes one PF to
@@ -401,7 +401,7 @@ in the ``options`` column of the ``Interface`` table.
    kernel netdevice, and be inherited from it when Open vSwitch is restarted,
    even if the options described in this section are unset from Open vSwitch.
 
-.. _bifurcated-drivers: https://doc.dpdk.org/guides-20.11/linux_gsg/linux_drivers.html#bifurcated-driver
+.. _bifurcated-drivers: https://doc.dpdk.org/guides-21.11/linux_gsg/linux_drivers.html#bifurcated-driver
 
 - Configure the VF MAC address::
 

Documentation/topics/dpdk/vdev.rst

@@ -63,4 +63,4 @@ run::
 More information on the different types of virtual DPDK PMDs can be found in
 the `DPDK documentation`__.
 
-__ https://doc.dpdk.org/guides-20.11/nics/overview.html
+__ https://doc.dpdk.org/guides-21.11/nics/overview.html

Documentation/topics/dpdk/vhost-user.rst

@@ -539,4 +539,4 @@ shown with::
 
 Further information can be found in the
 `DPDK documentation
-<https://doc.dpdk.org/guides-20.11/prog_guide/vhost_lib.html>`__
+<https://doc.dpdk.org/guides-21.11/prog_guide/vhost_lib.html>`__

Documentation/topics/testing.rst

@@ -353,7 +353,7 @@ All tests are skipped if no hugepages are configured. User must look into the DP
 manual to figure out how to `Configure hugepages`_.
 The phy test will skip if no compatible physical device is available.
 
-.. _Configure hugepages: https://doc.dpdk.org/guides-20.11/linux_gsg/sys_reqs.html
+.. _Configure hugepages: https://doc.dpdk.org/guides-21.11/linux_gsg/sys_reqs.html
 
 All the features documented under `Unit Tests`_ are available for the DPDK
 datapath testsuite.

Documentation/topics/userspace-tso.rst

@@ -46,7 +46,7 @@ datasheet for compatibility. Secondly, the NIC must have an associated DPDK
 Poll Mode Driver (PMD) which supports `TSO`. For a list of features per PMD,
 refer to the `DPDK documentation`__.
 
-__ https://doc.dpdk.org/guides-20.11/nics/overview.html
+__ https://doc.dpdk.org/guides-21.11/nics/overview.html
 
 Enabling TSO
 ~~~~~~~~~~~~

NEWS

@@ -10,6 +10,7 @@ Post-v2.16.0
        limiting behavior.
      * Add hardware offload support for matching IPv4/IPv6 frag types
        (experimental).
+     * Add support for DPDK 21.11.
    - Python:
      * For SSL support, the use of the pyOpenSSL library has been replaced
        with the native 'ssl' module.

lib/dp-packet.h

@@ -58,29 +58,31 @@ enum OVS_PACKED_ENUM dp_packet_source {
 enum dp_packet_offload_mask {
     /* Value 0 is not used. */
     /* Is the 'rss_hash' valid? */
-    DEF_OL_FLAG(DP_PACKET_OL_RSS_HASH, PKT_RX_RSS_HASH, 0x1),
+    DEF_OL_FLAG(DP_PACKET_OL_RSS_HASH, RTE_MBUF_F_RX_RSS_HASH, 0x1),
     /* Is the 'flow_mark' valid? */
-    DEF_OL_FLAG(DP_PACKET_OL_FLOW_MARK, PKT_RX_FDIR_ID, 0x2),
+    DEF_OL_FLAG(DP_PACKET_OL_FLOW_MARK, RTE_MBUF_F_RX_FDIR_ID, 0x2),
     /* Bad L4 checksum in the packet. */
-    DEF_OL_FLAG(DP_PACKET_OL_RX_L4_CKSUM_BAD, PKT_RX_L4_CKSUM_BAD, 0x4),
+    DEF_OL_FLAG(DP_PACKET_OL_RX_L4_CKSUM_BAD, RTE_MBUF_F_RX_L4_CKSUM_BAD, 0x4),
     /* Bad IP checksum in the packet. */
-    DEF_OL_FLAG(DP_PACKET_OL_RX_IP_CKSUM_BAD, PKT_RX_IP_CKSUM_BAD, 0x8),
+    DEF_OL_FLAG(DP_PACKET_OL_RX_IP_CKSUM_BAD, RTE_MBUF_F_RX_IP_CKSUM_BAD, 0x8),
     /* Valid L4 checksum in the packet. */
-    DEF_OL_FLAG(DP_PACKET_OL_RX_L4_CKSUM_GOOD, PKT_RX_L4_CKSUM_GOOD, 0x10),
+    DEF_OL_FLAG(DP_PACKET_OL_RX_L4_CKSUM_GOOD, RTE_MBUF_F_RX_L4_CKSUM_GOOD,
+                0x10),
     /* Valid IP checksum in the packet. */
-    DEF_OL_FLAG(DP_PACKET_OL_RX_IP_CKSUM_GOOD, PKT_RX_IP_CKSUM_GOOD, 0x20),
+    DEF_OL_FLAG(DP_PACKET_OL_RX_IP_CKSUM_GOOD, RTE_MBUF_F_RX_IP_CKSUM_GOOD,
+                0x20),
     /* TCP Segmentation Offload. */
-    DEF_OL_FLAG(DP_PACKET_OL_TX_TCP_SEG, PKT_TX_TCP_SEG, 0x40),
+    DEF_OL_FLAG(DP_PACKET_OL_TX_TCP_SEG, RTE_MBUF_F_TX_TCP_SEG, 0x40),
     /* Offloaded packet is IPv4. */
-    DEF_OL_FLAG(DP_PACKET_OL_TX_IPV4, PKT_TX_IPV4, 0x80),
+    DEF_OL_FLAG(DP_PACKET_OL_TX_IPV4, RTE_MBUF_F_TX_IPV4, 0x80),
     /* Offloaded packet is IPv6. */
-    DEF_OL_FLAG(DP_PACKET_OL_TX_IPV6, PKT_TX_IPV6, 0x100),
+    DEF_OL_FLAG(DP_PACKET_OL_TX_IPV6, RTE_MBUF_F_TX_IPV6, 0x100),
     /* Offload TCP checksum. */
-    DEF_OL_FLAG(DP_PACKET_OL_TX_TCP_CKSUM, PKT_TX_TCP_CKSUM, 0x200),
+    DEF_OL_FLAG(DP_PACKET_OL_TX_TCP_CKSUM, RTE_MBUF_F_TX_TCP_CKSUM, 0x200),
     /* Offload UDP checksum. */
-    DEF_OL_FLAG(DP_PACKET_OL_TX_UDP_CKSUM, PKT_TX_UDP_CKSUM, 0x400),
+    DEF_OL_FLAG(DP_PACKET_OL_TX_UDP_CKSUM, RTE_MBUF_F_TX_UDP_CKSUM, 0x400),
     /* Offload SCTP checksum. */
-    DEF_OL_FLAG(DP_PACKET_OL_TX_SCTP_CKSUM, PKT_TX_SCTP_CKSUM, 0x800),
+    DEF_OL_FLAG(DP_PACKET_OL_TX_SCTP_CKSUM, RTE_MBUF_F_TX_SCTP_CKSUM, 0x800),
     /* Adding new field requires adding to DP_PACKET_OL_SUPPORTED_MASK. */
 };
 

lib/netdev-dpdk.c

@@ -181,7 +181,7 @@ static int vring_state_changed(int vid, uint16_t queue_id, int enable);
 static void destroy_connection(int vid);
 static void vhost_guest_notified(int vid);
 
-static const struct vhost_device_ops virtio_net_device_ops =
+static const struct rte_vhost_device_ops virtio_net_device_ops =
 {
     .new_device =  new_device,
     .destroy_device = destroy_device,
@@ -1279,26 +1279,6 @@ common_construct(struct netdev *netdev, dpdk_port_t port_no,
     return 0;
 }
 
-/* Get the number of OVS interfaces which have the same DPDK
- * rte device (e.g. same pci bus address).
- * FIXME: avoid direct access to DPDK internal array rte_eth_devices.
- */
-static int
-netdev_dpdk_get_num_ports(struct rte_device *device)
-    OVS_REQUIRES(dpdk_mutex)
-{
-    struct netdev_dpdk *dev;
-    int count = 0;
-
-    LIST_FOR_EACH (dev, list_node, &dpdk_list) {
-        if (rte_eth_devices[dev->port_id].device == device
-        && rte_eth_devices[dev->port_id].state != RTE_ETH_DEV_UNUSED) {
-            count++;
-        }
-    }
-    return count;
-}
-
 static int
 vhost_common_construct(struct netdev *netdev)
     OVS_REQUIRES(dpdk_mutex)
@@ -1452,8 +1432,6 @@ static void
 netdev_dpdk_destruct(struct netdev *netdev)
 {
     struct netdev_dpdk *dev = netdev_dpdk_cast(netdev);
-    struct rte_device *rte_dev;
-    struct rte_eth_dev *eth_dev;
 
     ovs_mutex_lock(&dpdk_mutex);
 
@@ -1461,20 +1439,43 @@ netdev_dpdk_destruct(struct netdev *netdev)
     dev->started = false;
 
     if (dev->attached) {
-        /* Retrieve eth device data before closing it.
+        bool dpdk_resources_still_used = false;
-         * FIXME: avoid direct access to DPDK internal array rte_eth_devices.
+        struct rte_eth_dev_info dev_info;
-         */
+        dpdk_port_t sibling_port_id;
-        eth_dev = &rte_eth_devices[dev->port_id];
+
-        rte_dev = eth_dev->device;
+        /* Check if this netdev has siblings (i.e. shares DPDK resources) among
+         * other OVS netdevs. */
+        RTE_ETH_FOREACH_DEV_SIBLING (sibling_port_id, dev->port_id) {
+            struct netdev_dpdk *sibling;
+
+            /* RTE_ETH_FOREACH_DEV_SIBLING lists dev->port_id as part of the
+             * loop. */
+            if (sibling_port_id == dev->port_id) {
+                continue;
+            }
+            LIST_FOR_EACH (sibling, list_node, &dpdk_list) {
+                if (sibling->port_id != sibling_port_id) {
+                    continue;
+                }
+                dpdk_resources_still_used = true;
+                break;
+            }
+            if (dpdk_resources_still_used) {
+                break;
+            }
+        }
+
+        /* Retrieve eth device data before closing it. */
+        rte_eth_dev_info_get(dev->port_id, &dev_info);
 
         /* Remove the eth device. */
         rte_eth_dev_close(dev->port_id);
 
-        /* Remove this rte device and all its eth devices if all the eth
+        /* Remove the rte device if no associated eth device is used by OVS.
-         * devices belonging to the rte device are closed.
+         * Note: any remaining eth devices associated to this rte device are
-         */
+         * closed by DPDK ethdev layer. */
-        if (!netdev_dpdk_get_num_ports(rte_dev)) {
+        if (!dpdk_resources_still_used) {
-            int ret = rte_dev_remove(rte_dev);
+            int ret = rte_dev_remove(dev_info.device);
 
             if (ret < 0) {
                 VLOG_ERR("Device '%s' can not be detached: %s.",
@@ -2165,14 +2166,14 @@ netdev_dpdk_prep_hwol_packet(struct netdev_dpdk *dev, struct rte_mbuf *mbuf)
 {
     struct dp_packet *pkt = CONTAINER_OF(mbuf, struct dp_packet, mbuf);
 
-    if (mbuf->ol_flags & PKT_TX_L4_MASK) {
+    if (mbuf->ol_flags & RTE_MBUF_F_TX_L4_MASK) {
         mbuf->l2_len = (char *)dp_packet_l3(pkt) - (char *)dp_packet_eth(pkt);
         mbuf->l3_len = (char *)dp_packet_l4(pkt) - (char *)dp_packet_l3(pkt);
         mbuf->outer_l2_len = 0;
         mbuf->outer_l3_len = 0;
     }
 
-    if (mbuf->ol_flags & PKT_TX_TCP_SEG) {
+    if (mbuf->ol_flags & RTE_MBUF_F_TX_TCP_SEG) {
         struct tcp_header *th = dp_packet_l4(pkt);
 
         if (!th) {
@@ -2182,11 +2183,11 @@ netdev_dpdk_prep_hwol_packet(struct netdev_dpdk *dev, struct rte_mbuf *mbuf)
         }
 
         mbuf->l4_len = TCP_OFFSET(th->tcp_ctl) * 4;
-        mbuf->ol_flags |= PKT_TX_TCP_CKSUM;
+        mbuf->ol_flags |= RTE_MBUF_F_TX_TCP_CKSUM;
         mbuf->tso_segsz = dev->mtu - mbuf->l3_len - mbuf->l4_len;
 
-        if (mbuf->ol_flags & PKT_TX_IPV4) {
+        if (mbuf->ol_flags & RTE_MBUF_F_TX_IPV4) {
-            mbuf->ol_flags |= PKT_TX_IP_CKSUM;
+            mbuf->ol_flags |= RTE_MBUF_F_TX_IP_CKSUM;
         }
     }
     return true;
@@ -2533,7 +2534,7 @@ netdev_dpdk_filter_packet_len(struct netdev_dpdk *dev, struct rte_mbuf **pkts,
     for (i = 0; i < pkt_cnt; i++) {
         pkt = pkts[i];
         if (OVS_UNLIKELY((pkt->pkt_len > dev->max_packet_len)
-            && !(pkt->ol_flags & PKT_TX_TCP_SEG))) {
+            && !(pkt->ol_flags & RTE_MBUF_F_TX_TCP_SEG))) {
             VLOG_WARN_RL(&rl, "%s: Too big size %" PRIu32 " "
                          "max_packet_len %d", dev->up.name, pkt->pkt_len,
                          dev->max_packet_len);
@@ -2755,12 +2756,12 @@ dpdk_copy_dp_packet_to_mbuf(struct rte_mempool *mp, struct dp_packet *pkt_orig)
     mbuf_dest->tx_offload = pkt_orig->mbuf.tx_offload;
     mbuf_dest->packet_type = pkt_orig->mbuf.packet_type;
     mbuf_dest->ol_flags |= (pkt_orig->mbuf.ol_flags &
-                            ~(EXT_ATTACHED_MBUF | IND_ATTACHED_MBUF));
+                            ~(RTE_MBUF_F_EXTERNAL | RTE_MBUF_F_INDIRECT));
 
     memcpy(&pkt_dest->l2_pad_size, &pkt_orig->l2_pad_size,
            sizeof(struct dp_packet) - offsetof(struct dp_packet, l2_pad_size));
 
-    if (mbuf_dest->ol_flags & PKT_TX_L4_MASK) {
+    if (mbuf_dest->ol_flags & RTE_MBUF_F_TX_L4_MASK) {
         mbuf_dest->l2_len = (char *)dp_packet_l3(pkt_dest)
                                 - (char *)dp_packet_eth(pkt_dest);
         mbuf_dest->l3_len = (char *)dp_packet_l4(pkt_dest)
@@ -2805,7 +2806,7 @@ dpdk_do_tx_copy(struct netdev *netdev, int qid, struct dp_packet_batch *batch)
         uint32_t size = dp_packet_size(packet);
 
         if (size > dev->max_packet_len
-            && !(packet->mbuf.ol_flags & PKT_TX_TCP_SEG)) {
+            && !(packet->mbuf.ol_flags & RTE_MBUF_F_TX_TCP_SEG)) {
             VLOG_WARN_RL(&rl, "Too big size %u max_packet_len %d", size,
                          dev->max_packet_len);
             mtu_drops++;
@@ -3679,8 +3680,8 @@ netdev_dpdk_get_status(const struct netdev *netdev, struct smap *args)
     ovs_mutex_unlock(&dev->mutex);
     const struct rte_bus *bus;
     const struct rte_pci_device *pci_dev;
-    uint16_t vendor_id = PCI_ANY_ID;
+    uint16_t vendor_id = RTE_PCI_ANY_ID;
-    uint16_t device_id = PCI_ANY_ID;
+    uint16_t device_id = RTE_PCI_ANY_ID;
     bus = rte_bus_find_by_device(dev_info.device);
     if (bus && !strcmp(bus->name, "pci")) {
         pci_dev = RTE_DEV_TO_PCI(dev_info.device);
@@ -3793,12 +3794,12 @@ static void
 netdev_dpdk_detach(struct unixctl_conn *conn, int argc OVS_UNUSED,
                    const char *argv[], void *aux OVS_UNUSED)
 {
-    char *response;
-    dpdk_port_t port_id;
-    struct netdev_dpdk *dev;
-    struct rte_device *rte_dev;
     struct ds used_interfaces = DS_EMPTY_INITIALIZER;
+    struct rte_eth_dev_info dev_info;
+    dpdk_port_t sibling_port_id;
+    dpdk_port_t port_id;
     bool used = false;
+    char *response;
 
     ovs_mutex_lock(&dpdk_mutex);
 
@@ -3808,18 +3809,21 @@ netdev_dpdk_detach(struct unixctl_conn *conn, int argc OVS_UNUSED,
         goto error;
     }
 
-    rte_dev = rte_eth_devices[port_id].device;
     ds_put_format(&used_interfaces,
                   "Device '%s' is being used by the following interfaces:",
                   argv[1]);
 
-    LIST_FOR_EACH (dev, list_node, &dpdk_list) {
+    RTE_ETH_FOREACH_DEV_SIBLING (sibling_port_id, port_id) {
-        /* FIXME: avoid direct access to DPDK array rte_eth_devices. */
+        struct netdev_dpdk *dev;
-        if (rte_eth_devices[dev->port_id].device == rte_dev
+
-            && rte_eth_devices[dev->port_id].state != RTE_ETH_DEV_UNUSED) {
+        LIST_FOR_EACH (dev, list_node, &dpdk_list) {
+            if (dev->port_id != sibling_port_id) {
+                continue;
+            }
             used = true;
             ds_put_format(&used_interfaces, " %s",
                           netdev_get_name(&dev->up));
+            break;
         }
     }
 
@@ -3831,8 +3835,9 @@ netdev_dpdk_detach(struct unixctl_conn *conn, int argc OVS_UNUSED,
     }
     ds_destroy(&used_interfaces);
 
+    rte_eth_dev_info_get(port_id, &dev_info);
     rte_eth_dev_close(port_id);
-    if (rte_dev_remove(rte_dev) < 0) {
+    if (rte_dev_remove(dev_info.device) < 0) {
         response = xasprintf("Device '%s' can not be detached", argv[1]);
         goto error;
     }
@@ -5272,7 +5277,7 @@ netdev_dpdk_rte_flow_query_count(struct netdev *netdev,
                                  struct rte_flow_query_count *query,
                                  struct rte_flow_error *error)
 {
-    struct rte_flow_action_count count = { .shared = 0, .id = 0 };
+    struct rte_flow_action_count count = { .id = 0, };
     const struct rte_flow_action actions[] = {
         {
             .type = RTE_FLOW_ACTION_TYPE_COUNT,

rhel/openvswitch-fedora.spec.in

@@ -74,7 +74,7 @@ BuildRequires: libcap-ng libcap-ng-devel
 %endif
 %if %{with dpdk}
 BuildRequires: libpcap-devel numactl-devel
-BuildRequires: dpdk-devel >= 20.11
+BuildRequires: dpdk-devel >= 21.11
 Provides: %{name}-dpdk = %{version}-%{release}
 %endif
 %if %{with afxdp}

tests/system-dpdk.at

@@ -15,7 +15,8 @@ AT_CHECK([grep "DPDK Enabled - initialized" ovs-vswitchd.log], [], [stdout])
 OVS_VSWITCHD_STOP(["/Global register is changed during/d
 /EAL:   Invalid NUMA socket, default to 0/d
 /EAL: WARNING: cpu flags constant_tsc=yes nonstop_tsc=no -> using unreliable clock cycles !/d
-/EAL: No free hugepages reported in hugepages-1048576kB/d"])
+/EAL: No free hugepages reported in hugepages-1048576kB/d
+/TELEMETRY: No legacy callbacks, legacy socket not created/d"])
 AT_CLEANUP
 dnl --------------------------------------------------------------------------
 
@@ -37,12 +38,12 @@ sleep 2
 
 dnl Clean up
 AT_CHECK([ovs-vsctl del-port br10 phy0], [], [stdout], [stderr])
-OVS_VSWITCHD_STOP("/does not exist. The Open vSwitch kernel module is probably not loaded./d
+OVS_VSWITCHD_STOP(["/does not exist. The Open vSwitch kernel module is probably not loaded./d
 /Failed to enable flow control/d
 /Global register is changed during/d
 /EAL: WARNING: cpu flags constant_tsc=yes nonstop_tsc=no -> using unreliable clock cycles !/d
 /EAL: No free hugepages reported in hugepages-1048576kB/d
-")
+/TELEMETRY: No legacy callbacks, legacy socket not created/d"])
 AT_CLEANUP
 dnl --------------------------------------------------------------------------
 
@@ -74,7 +75,8 @@ OVS_VSWITCHD_STOP(["\@does not exist. The Open vSwitch kernel module is probably
 \@Global register is changed during@d
 \@EAL:   Invalid NUMA socket, default to 0@d
 \@EAL: WARNING: cpu flags constant_tsc=yes nonstop_tsc=no -> using unreliable clock cycles !@d
-\@EAL: No free hugepages reported in hugepages-1048576kB@d"])
+\@EAL: No free hugepages reported in hugepages-1048576kB@d
+\@TELEMETRY: No legacy callbacks, legacy socket not created@d"])
 AT_CLEANUP
 dnl --------------------------------------------------------------------------
 
@@ -152,7 +154,8 @@ OVS_VSWITCHD_STOP(["\@does not exist. The Open vSwitch kernel module is probably
 \@failed to enumerate system datapaths: No such file or directory@d
 \@EAL:   Invalid NUMA socket, default to 0@d
 \@EAL: WARNING: cpu flags constant_tsc=yes nonstop_tsc=no -> using unreliable clock cycles !@d
-\@EAL: No free hugepages reported in hugepages-1048576kB@d"])
+\@EAL: No free hugepages reported in hugepages-1048576kB@d
+\@TELEMETRY: No legacy callbacks, legacy socket not created@d"])
 AT_CLEANUP
 dnl --------------------------------------------------------------------------
 
@@ -227,7 +230,8 @@ OVS_VSWITCHD_STOP(["\@does not exist. The Open vSwitch kernel module is probably
 \@failed to enumerate system datapaths: No such file or directory@d
 \@EAL:   Invalid NUMA socket, default to 0@d
 \@EAL: WARNING: cpu flags constant_tsc=yes nonstop_tsc=no -> using unreliable clock cycles !@d
-\@EAL: No free hugepages reported in hugepages-1048576kB@d"])
+\@EAL: No free hugepages reported in hugepages-1048576kB@d
+\@TELEMETRY: No legacy callbacks, legacy socket not created@d"])
 AT_CLEANUP
 dnl --------------------------------------------------------------------------