iavf: Fix error when changing ring parameters on ice PF
Subsystems (2)
PATCH v4 1/1 iavf: Fix error when changing ring parameters on ice PF
Reset is triggered when ring parameters are being changed through ethtool and queues are reconfigured for VF's VSI. If ring is changed again immediately, then the next reset could be executed before queues could be properly reinitialized on VF's VSI. It caused ice PF to mess up the VSI resource tree.
Add a check in iavf_set_ringparam for adapter and VF's queue state. If VF is currently resetting or queues are disabled for the VF return with EAGAIN error.
---
v4: Changed order of if statements suggested by community
Answer to question:
Can't we wait for the device to get into the right state? Throwing EAGAIN back to user space is not very friendly. Answer:
When we have state which is running, it does not mean that we have queues configured on PF. So in order to configure queues on PF, the IAVF_FLAG_QUEUES has to be disabled. I use here EAGAIN because as long as we are not configured with queues, it does not make any sens to trigger command and we are not sure when the configuration of queues will end - so that is why EAGAIN is used.
v3: Correct patch to be send on right tree
v2: Changed unnecessary parentheses
---
--- drivers/net/ethernet/intel/iavf/iavf_ethtool.c | 5 +++++ 1 file changed, 5 insertions(+)
drivers/net/ethernet/intel/iavf/iavf_ethtool.c +5 -0
@@ -628,6 +628,11 @@ static int iavf_set_ringparam(struct net_device *netdev,struct iavf_adapter *adapter =netdev_priv (netdev);u32 new_rx_count, new_tx_count; +if (adapter->state == __IAVF_RESETTING || + adapter->state == __IAVF_RUNNING && + (adapter->flags &IAVF_FLAG_QUEUES_DISABLED )) +return -EAGAIN ; +if ((ring->rx_mini_pending) || (ring->rx_jumbo_pending))return -EINVAL ;
Signature · 1 line
2.27.0