Restore the call to idpf_vc_xn_shutdown() at the beginning of
idpf_vc_core_deinit() provided the function is not called on remove.
In the reset path the mailbox is destroyed, leading to all transactions
timing out.
Fixes: 09d0fb5cb30e ("idpf: deinit virtchnl transaction manager after vport and
vectors")
Reviewed-by: Larysa Zaremba <[email protected]>
Signed-off-by: Emil Tantilov <[email protected]>
---
Changelog:
v2:
- Assigned the current state of REMOVE_IN_PROG flag to a boolean
variable, to be checked instead of reading the flag twice.
- Updated the description to clarify the reason for the timeouts on
reset is due to the mailbox being destroyed.
v1:
https://lore.kernel.org/intel-wired-lan/[email protected]/
Testing hints:
echo 1 > /sys/class/net/<netif>/device/reset
---
drivers/net/ethernet/intel/idpf/idpf_virtchnl.c | 11 ++++++++++-
1 file changed, 10 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c
b/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c
index d46c95f91b0d..7639d520b806 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c
@@ -3077,12 +3077,21 @@ int idpf_vc_core_init(struct idpf_adapter *adapter)
*/
void idpf_vc_core_deinit(struct idpf_adapter *adapter)
{
+ bool remove_in_prog;
+
if (!test_bit(IDPF_VC_CORE_INIT, adapter->flags))
return;
+ /* Avoid transaction timeouts when called during reset */
+ remove_in_prog = test_bit(IDPF_REMOVE_IN_PROG, adapter->flags);
+ if (!remove_in_prog)
+ idpf_vc_xn_shutdown(adapter->vcxn_mngr);
+
idpf_deinit_task(adapter);
idpf_intr_rel(adapter);
- idpf_vc_xn_shutdown(adapter->vcxn_mngr);
+
+ if (remove_in_prog)
+ idpf_vc_xn_shutdown(adapter->vcxn_mngr);
cancel_delayed_work_sync(&adapter->serv_task);
cancel_delayed_work_sync(&adapter->mbx_task);
--
2.17.2
