29/09/2021 16:52, dkozl...@oss.nvidia.com:
> From: Dmitry Kozlyuk <dkozl...@oss.nvidia.com>
>
> Performance of MLX5 PMD of different classes can benefit if PMD knows
> which memory it will need to handle in advance, before the first mbuf
> is sent to the PMD. It is impractical, however, to consider
> all allocated memory for this purpose. Most often mbuf memory comes
> from mempools that can come and go. PMD can enumerate existing mempools
> on device start, but it also needs to track creation and destruction
> of mempools after the forwarding starts but before an mbuf from the new
> mempool is sent to the device.
I'm not sure this introduction about mlx5 is appropriate.
> Add an internal API to register callback for mempool lify cycle events,
lify -> life
> currently RTE_MEMPOOL_EVENT_READY (after populating)
> and RTE_MEMPOOL_EVENT_DESTROY (before freeing):
> * rte_mempool_event_callback_register()
> * rte_mempool_event_callback_unregister()
> Provide a unit test for the new API.
[...]
> -rte_mempool_create_empty(const char *name, unsigned n, unsigned elt_size,
> - unsigned cache_size, unsigned private_data_size,
> - int socket_id, unsigned flags)
> +rte_mempool_create_empty(const char *name, unsigned int n,
> + unsigned int elt_size, unsigned int cache_size,
> + unsigned int private_data_size, int socket_id, unsigned int flags)
This change looks unrelated.
> +enum rte_mempool_event {
> + /** Occurs after a mempool is successfully populated. */
> + RTE_MEMPOOL_EVENT_READY = 0,
> + /** Occurs before destruction of a mempool begins. */
> + RTE_MEMPOOL_EVENT_DESTROY = 1,
> +};
These events look OK.
> +typedef void (rte_mempool_event_callback)(
> + enum rte_mempool_event event,
> + struct rte_mempool *mp,
> + void *arg);
Instead of "arg", I prefer the name "user_data".
