All of lore.kernel.org
 help / color / mirror / Atom feed
* [DPDK/ethdev Bug 1900] The DPDK bond driver does not support the secondary process.
@ 2026-03-10  2:53 bugzilla
  0 siblings, 0 replies; only message in thread
From: bugzilla @ 2026-03-10  2:53 UTC (permalink / raw)
  To: dev

http://bugs.dpdk.org/show_bug.cgi?id=1900

            Bug ID: 1900
           Summary: The DPDK bond driver does not support the secondary
                    process.
           Product: DPDK
           Version: unspecified
          Hardware: All
                OS: All
            Status: UNCONFIRMED
          Severity: normal
          Priority: Normal
         Component: ethdev
          Assignee: dev@dpdk.org
          Reporter: 916421004@qq.com
  Target Milestone: ---

When I use the bond4 mode (dedicated_queues_enable), I find that DPDK does not
support the secondary mode.


1.bond_probe:
if (rte_eal_process_type() == RTE_PROC_SECONDARY) {
        eth_dev = rte_eth_dev_attach_secondary(name);
        if (!eth_dev) {
                RTE_BOND_LOG(ERR, "Failed to probe %s", name);
                return -1;
        }
        /* TODO: request info from primary to set up Rx and Tx */
        eth_dev->dev_ops = &default_dev_ops;
        eth_dev->device = &dev->device;

        rte_eth_dev_probing_finish(eth_dev);
        return 0;
}


Here don`t set eth_dev->tx_pkt_burst and eth_dev->rx_pkt_burst.

2.struct port bond_mode_8023ad_ports[RTE_MAX_ETHPORTS];
This is a global variable, and the content of the primary is not synchronized.

problem example:
rx_burst_8023ad:
...
collecting = ACTOR_STATE(&bond_mode_8023ad_ports[members[idx]],
                                         COLLECTING);
...                     
/* Remove packet from array if:
 * - it is slow packet but no dedicated rxq is present,
 * - member is not in collecting state,
 * - bonding interface is not in promiscuous mode and
 *   packet address isn't in mac_addrs array:
 *   - packet is unicast,
 *   - packet is multicast and bonding interface
 *     is not in allmulti,
 */
if (unlikely(
        (!dedicated_rxq &&
         is_lacp_packets(hdr->ether_type, subtype,
                         bufs[j])) ||
        !collecting ||
        (!promisc &&
         !is_bond_mac_addr(&hdr->dst_addr, bond_mac,
                           BOND_MAX_MAC_ADDRS) &&
         (rte_is_unicast_ether_addr(&hdr->dst_addr) ||
          !allmulti)))) {
          ...
          }

because the bond_mode_8023ad_ports is not synchronized to the primary process.
As a result, the value of collecting is 0, and non-LACP packets are discarded.


Why does the DPDK bond not support the secondary mode? When will the DPDK bond
support the secondary mode?

-- 
You are receiving this mail because:
You are the assignee for the bug.

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2026-03-10  2:53 UTC | newest]

Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-03-10  2:53 [DPDK/ethdev Bug 1900] The DPDK bond driver does not support the secondary process bugzilla

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.