Resend trap 144 when detecting remote MASTER SM with lower priority
PATCH 1/1 Resend trap 144 when detecting remote MASTER SM with lower priority
From: Alex Netes <alexne-VPRAkNaXOzVWk0Htik3J/[email protected]>
Remote MASTER SM might not get trap 144 and therefore never perform a HANDOVER. The patch solves this by resending trap 144 when receiving GetResp(SMInfo) to a remote MASTER SM after polling timeout expires.
--- opensm/osm_sminfo_rcv.c | 17 +++++++++++++---- 1 files changed, 13 insertions(+), 4 deletions(-)
opensm/osm_sminfo_rcv.c +13 -4
@@ -383,11 +383,20 @@ static void smi_rcv_process_get_sm(IN osm_sm_t * sm,switch (ib_sminfo_get_state (p_smi)) {case IB_SMINFO_STATE_MASTER :/* If this is a response due to our polling, this means that we are -waiting for a handover from this SM, and it is still alive - -signal that. */ -if (sm->p_polling_sm) +* waiting for a handover from this SM, and it is still alive - +* signal that. If we detected the remote SM with higher priority +* we should init a heavy sweep in order to go STANDBY. If we +* detected a remote SM with lower priority, we should resend trap144 +* as it might not get it and we don't want to wait for a HANDOVER +* forever. +*/ +if (sm->p_polling_sm) { +if (smi_rcv_remote_sm_is_higher (sm, p_smi)) + sm->p_subn->force_heavy_sweep = TRUE; +else +osm_send_trap144 (sm,TRAP_144_MASK_SM_PRIORITY_CHANGE );osm_sm_state_mgr_signal_master_is_alive (sm); -else { + }else {/* This is a response we got while sweeping the subnet. * * If this is during a heavy sweep, we will handle a case of
Signature · 6 lines
1.7.8.2 -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html