[dpdk-dev] [PATCH 1/3] ethdev: add new eth_dev_ops function for mtr ops get

[Jerin Jacob]

-----Original Message-----
> Date: Sat, 26 Aug 2017 01:06:11 +0100
> From: Cristian Dumitrescu <cristian.dumitrescu at intel.com>
> To: dev at dpdk.org
> CC: thomas at monjalon.net, adrien.mazarguil at 6wind.com,
>  jerin.jacob at caviumnetworks.com, hemant.agrawal at nxp.com
> Subject: [PATCH 1/3] ethdev: add new eth_dev_ops function for mtr ops get
> X-Mailer: git-send-email 2.7.4
> 
> Following similar approach to rte_flow and rte_tm for modularity reasons.
> 
> Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu at intel.com>
> ---
> Changes in v1 (from RFC [1]):
> - Removed ethdev API function to get the MTR ops, as it is not needed
>   (input from Thomas)
> 
> [1] RFC: http://www.dpdk.org/ml/archives/dev/2017-May/066888.html
> 
>  lib/librte_ether/rte_ethdev.h | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/lib/librte_ether/rte_ethdev.h b/lib/librte_ether/rte_ethdev.h
> index 0adf327..361c9bd 100644
> --- a/lib/librte_ether/rte_ethdev.h
> +++ b/lib/librte_ether/rte_ethdev.h
> @@ -1421,6 +1421,9 @@ typedef int (*eth_filter_ctrl_t)(struct rte_eth_dev *dev,
>  typedef int (*eth_tm_ops_get_t)(struct rte_eth_dev *dev, void *ops);
>  /**< @internal Get Traffic Management (TM) operations on an Ethernet device */
>  
> +typedef int (*eth_mtr_ops_get_t)(struct rte_eth_dev *dev, void *ops);
> +/**< @internal Get Trafffic Metering and Policing (MTR) operations */
> +
>  typedef int (*eth_get_dcb_info)(struct rte_eth_dev *dev,
>  				 struct rte_eth_dcb_info *dcb_info);
>  /**< @internal Get dcb information on an Ethernet device */
> @@ -1544,6 +1547,9 @@ struct eth_dev_ops {
>  
>  	eth_tm_ops_get_t tm_ops_get;
>  	/**< Get Traffic Management (TM) operations. */
> +
> +	eth_mtr_ops_get_t mtr_ops_get;
> +	/**< Get Traffic Metering and Policing (MTR) operations. */

It will useful to have capability info here. In Cavium hardwares,
all input pkts are considered as green when they enqueue to ethdev tx
queue.

On the congestion control on Rx side, We support tail drop and RED per
mempool(not per RX queue)