21/06/2022 03:38, fengchengwen:
> Hi Thomas,
> 
> On 2022/6/21 1:42, Thomas Monjalon wrote:
> > 16/06/2022 11:41, Chengwen Feng:
> >> --- a/lib/ethdev/rte_ethdev.h
> >> +++ b/lib/ethdev/rte_ethdev.h
> >> @@ -3928,6 +3928,12 @@ enum rte_eth_event_type {
> >>     * @see rte_eth_rx_avail_thresh_set()
> >>     */
> >>    RTE_ETH_EVENT_RX_AVAIL_THRESH,
> >> +  /** Port recovering from a hardware or firmware error */
> >> +  RTE_ETH_EVENT_ERR_RECOVERING,
> >> +  /** Port recovers successful from the error */
> >> +  RTE_ETH_EVENT_RECOVER_SUCCESS,
> >> +  /** Port recovers failed from the error */
> >> +  RTE_ETH_EVENT_RECOVER_FAILED,
> >>    RTE_ETH_EVENT_MAX       /**< max value of this enum */
> >>  };
> > 
> > The descriptions here are not enough.
> > We cannot understand what has changed on the port,
> > and which action must be taken.
> 
> There are detail descriptions in /doc/guides/prog_guide/poll_mode_drv.rst,
> I will add your review in poll_mode_drv.rst.
> 
> Another question: do we need to add a detail description here as well? I 
> think the poll_mode_drv.rst is enough.

It is the opposite: the RST guide is to give the overview,
while the doxygen comments are the precise API documentation.
You need to explain what is the state of the device.
Is it the same as after a call to rte_eth_dev_stop() ?
The application needs to know what must be reconfigured.



Reply via email to