[LU-13277] Potential deadlock in lnet_peer_discovery

Thu, 5 Mar 2020 23:26:23 +0000

Potential deadlock here when LNet is shutting down:

static int lnet_peer_discovery(void *arg)
...
        for (;;) {
...
                lnet_net_lock(LNET_LOCK_EX);
                if (the_lnet.ln_dc_state == LNET_DC_STATE_STOPPING)
                        break;
...
        }

        CDEBUG(D_NET, "stopping\n");
        /*
         * Clean up before telling lnet_peer_discovery_stop() that
         * we're done. Use wake_up() below to somewhat reduce the
         * size of the thundering herd if there are multiple threads
         * waiting on discovery of a single peer.
         */

        /* Queue cleanup 1: stop all pending pings and pushes. */
        lnet_net_lock(LNET_LOCK_EX); <<< Deadlock
...

Whamcloud Community JIRA

[LU-13277] Potential deadlock in lnet_peer_discovery