[Intel-wired-lan] [PATCH] igb: reinit_locked() should be called with rtnl_lock
Francesco Ruggeri
fruggeri at arista.com
Mon Jun 29 21:18:01 UTC 2020
We observed a panic in igb_reset_task caused by this race condition
when doing a reboot -f:
kworker reboot -f
igb_reset_task
igb_reinit_locked
igb_down
napi_synchronize
__igb_shutdown
igb_clear_interrupt_scheme
igb_free_q_vectors
igb_free_q_vector
adapter->q_vector[v_idx] = NULL;
napi_disable
Panics trying to access
adapter->q_vector[v_idx].napi_state
This commit applies to igb the same changes that were applied to ixgbe
in commit 8f4c5c9fb87a ("ixgbe: reinit_locked() should be called with
rtnl_lock") and commit 88adce4ea8f9 ("ixgbe: fix possible race in
reset subtask").
Signed-off-by: Francesco Ruggeri <fruggeri at arista.com>
---
drivers/net/ethernet/intel/igb/igb_main.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 8bb3db2cbd41..b79a78e102f3 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -6224,9 +6224,11 @@ static void igb_reset_task(struct work_struct *work)
struct igb_adapter *adapter;
adapter = container_of(work, struct igb_adapter, reset_task);
+ rtnl_lock();
igb_dump(adapter);
netdev_err(adapter->netdev, "Reset adapter\n");
igb_reinit_locked(adapter);
+ rtnl_unlock();
}
/**
More information about the Intel-wired-lan
mailing list