[dpdk-dev] [Bug 52] Bonding PMD may fail to accept new slaves in certain conditions

[dpdk-dev] [Bug 52] Bonding PMD may fail to accept new slaves in certain conditions

[bugzilla]

https://dpdk.org/tracker/show_bug.cgi?id=52

            Bug ID: 52
           Summary: Bonding PMD may fail to accept new slaves in certain
                    conditions
           Product: DPDK
           Version: 18.05
          Hardware: All
                OS: All
            Status: CONFIRMED
          Severity: minor
          Priority: Normal
         Component: ethdev
          Assignee: dev@dpdk.org
          Reporter: radu.nicolau@intel.com
  Target Milestone: ---

When using testpmd to test bonding, the following sequence of commands will
fail as described:

testpmd> port stop all
testpmd> create bonded device 4 0
testpmd> add bonding slave 0 2
testpmd> add bonding slave 1 2

__eth_bond_slave_add_lock_free(352) - Invalid link properties for slave 1 in
bonding mode 4
Failed to add slave 1 to master port = 2


Root cause is a mismatch in port link status caused by testpmd implementation
that does not re-check all ports, but stops at the first port that shows link
down.

To work around the issue insert "show port info all" command after "port stop
all" if using testpmd, or make sure that the application either calls
rte_eth_link_get on all ports that will be bound, or on none of them.

Patch addressing the issue: https://dpdk.org/dev/patchwork/patch/39567/

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