Hi M. Bot,
See below
quoted
+void dpaa2_eth_dl_port_check(struct dpaa2_eth_priv *priv)
+{
+ u32 registered = priv->devlink_port.attrs.phys.port_number;
+ u32 current_number = dpaa2_eth_dl_port_number(priv);
[Severity: Low]
Can this read of priv->devlink_port.attrs race with probe and remove?
Yes, see v2. I did add the port number in priv under mac_lock.
[Severity: Low]
Does the "current" value used by dpaa2_eth_dl_port_check() really
reflect the firmware endpoint?
Yes, it does. The 0: non was missleading.
The prints is updated following the reiews and Iona's comments, see v2.
[Severity: High]
This is a pre-existing issue and was not introduced by this patch, but
the new dpaa2_eth_dl_port_check() call adds another priv access on this
path. Can dpni_irq0_handler_thread() run after priv has been freed?
I could be, it was there before, so let's avoid unfocusing this serie.
Best regards,
Vincent