Commit 33006858 authored by Emil Tantilov's avatar Emil Tantilov Committed by Tony Nguyen
Browse files

idpf: disable local BH when scheduling napi for marker packets 



Fix softirq's not being handled during napi_schedule() call when
receiving marker packets for queue disable by disabling local bottom
half.

The issue can be seen on ifdown:
NOHZ tick-stop error: Non-RCU local softirq work is pending, handler #08!!!

Using ftrace to catch the failing scenario:
ifconfig   [003] d.... 22739.830624: softirq_raise: vec=3 [action=NET_RX]
<idle>-0   [003] ..s.. 22739.831357: softirq_entry: vec=3 [action=NET_RX]

No interrupt and CPU is idle.

After the patch when disabling local BH before calling napi_schedule:
ifconfig   [003] d.... 22993.928336: softirq_raise: vec=3 [action=NET_RX]
ifconfig   [003] ..s1. 22993.928337: softirq_entry: vec=3 [action=NET_RX]

Fixes: c2d548ca ("idpf: add TX splitq napi poll support")
Reviewed-by: default avatarJesse Brandeburg <jesse.brandeburg@intel.com>
Reviewed-by: default avatarPrzemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: default avatarEmil Tantilov <emil.s.tantilov@intel.com>
Signed-off-by: default avatarAlan Brady <alan.brady@intel.com>
Reviewed-by: default avatarSimon Horman <horms@kernel.org>
Tested-by: default avatarKrishneil Singh <krishneil.k.singh@intel.com>
Signed-off-by: default avatarTony Nguyen <anthony.l.nguyen@intel.com>
parent 4daa8731

drivers/net/ethernet/intel/idpf/idpf_virtchnl.c

+2 −0
Original line number Diff line number Diff line
@@ -2087,8 +2087,10 @@ int idpf_send_disable_queues_msg(struct idpf_vport *vport) 
		set_bit(__IDPF_Q_POLL_MODE, vport->txqs[i]->flags);



	/* schedule the napi to receive all the marker packets */

	local_bh_disable();

	for (i = 0; i < vport->num_q_vectors; i++)

		napi_schedule(&vport->q_vectors[i].napi);

	local_bh_enable();



	return idpf_wait_for_marker_event(vport);

}